
From fred@cisco.com  Wed Jan  1 22:31:30 2014
Return-Path: <fred@cisco.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 38FD11AE142 for <aqm@ietfa.amsl.com>; Wed,  1 Jan 2014 22:31:30 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -110.039
X-Spam-Level: 
X-Spam-Status: No, score=-110.039 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RP_MATCHES_RCVD=-0.538, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5, USER_IN_WHITELIST=-100] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Jka1MoiA9cuO for <aqm@ietfa.amsl.com>; Wed,  1 Jan 2014 22:31:28 -0800 (PST)
Received: from alln-iport-8.cisco.com (alln-iport-8.cisco.com [173.37.142.95]) by ietfa.amsl.com (Postfix) with ESMTP id A62061AE2DB for <aqm@ietf.org>; Wed,  1 Jan 2014 22:31:28 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=1469; q=dns/txt; s=iport; t=1388644282; x=1389853882; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=kq7V5L9SMY+zai9cmYQ2rpyYzCvOpFjPEEMhIuujjLM=; b=jHZWipu9P1uBZ1z2FJvYEZKQ/GQyRooHCLHDo08wSK75HnwY9ieJAH46 uXtJGecd89ZS/lWr/eNSu8NpImQ58APminY1g1xLhGouw/4SEIW1rKR+k 1FWgzUldfR/+RSpfV5cr9xA8HFdEFH575DxogjhXR4crlioNtky3WZFZ+ A=;
X-Files: signature.asc : 195
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: AgUFAPoGxVKtJXG9/2dsb2JhbABYgwuBDadbkReBDRZ0giUBAQEDAXkFCwIBCEYyJQIECgQFDoduCMIqF48dB4MjgRMEkDOBMYYzkhSDLYIq
X-IronPort-AV: E=Sophos;i="4.95,589,1384300800";  d="asc'?scan'208";a="10049837"
Received: from rcdn-core2-2.cisco.com ([173.37.113.189]) by alln-iport-8.cisco.com with ESMTP; 02 Jan 2014 06:31:21 +0000
Received: from xhc-rcd-x06.cisco.com (xhc-rcd-x06.cisco.com [173.37.183.80]) by rcdn-core2-2.cisco.com (8.14.5/8.14.5) with ESMTP id s026VL0d015130 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 2 Jan 2014 06:31:21 GMT
Received: from xmb-rcd-x09.cisco.com ([169.254.9.86]) by xhc-rcd-x06.cisco.com ([173.37.183.80]) with mapi id 14.03.0123.003; Thu, 2 Jan 2014 00:31:21 -0600
From: "Fred Baker (fred)" <fred@cisco.com>
To: "<curtis@ipv6.occnc.com>" <curtis@ipv6.occnc.com>
Thread-Topic: [aqm] What is a good burst? -- AQM evaluation guidelines
Thread-Index: AQHPB4Q//OfwR1C28UGyjPKOXs6g3g==
Date: Thu, 2 Jan 2014 06:31:20 +0000
Message-ID: <533BE7A9-7804-4A74-BBFB-C75CCE212434@cisco.com>
References: <201312151857.rBFIuuea043478@gateway0.ipv6.occnc.com>
In-Reply-To: <201312151857.rBFIuuea043478@gateway0.ipv6.occnc.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
x-originating-ip: [10.19.64.120]
Content-Type: multipart/signed; boundary="Apple-Mail=_E46BB7BE-5C7F-47D1-ADCE-2633CEAEE2E4"; protocol="application/pgp-signature"; micalg=pgp-sha1
MIME-Version: 1.0
Cc: bloat <bloat@lists.bufferbloat.net>, "aqm@ietf.org" <aqm@ietf.org>, Naeem Khademi <naeem.khademi@gmail.com>, "<end2end-interest@postel.org>" <end2end-interest@postel.org>
Subject: Re: [aqm] What is a good burst? -- AQM evaluation guidelines
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 02 Jan 2014 06:31:30 -0000

--Apple-Mail=_E46BB7BE-5C7F-47D1-ADCE-2633CEAEE2E4
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii


On Dec 15, 2013, at 10:56 AM, Curtis Villamizar <curtis@ipv6.occnc.com> =
wrote:

> So briefly, my answer is: as a WG, I don't think we want to go there.
> If we do go there at all, then we should define "good AQM" in terms of
> acheving a "good" tradeoff between fairness, bulk transfer goodput,
> and bounded delay.  IMHO sometimes vague is better.

As you may have worked out from my previous comments in these threads, I =
agree with you. I don't think this can be nailed down in a universal =
sense. What can be described is the result in the network, in that =
delays build up that persist, as opposed to coming and going, and as a =
result applications don't work as well as they might - and at that =
point, it is appropriate for the network to inform the transport.

--Apple-Mail=_E46BB7BE-5C7F-47D1-ADCE-2633CEAEE2E4
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment; filename="signature.asc"
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Message signed with OpenPGP using GPGMail

-----BEGIN PGP SIGNATURE-----
Comment: GPGTools - http://gpgtools.org

iD8DBQFSxQe2bjEdbHIsm0MRAuSyAKCWM1N4nXUaneofDIU/9TrRfLoZBwCeJvmV
JtT1oLqhbYH9e11SzOcuhIg=
=4pgW
-----END PGP SIGNATURE-----

--Apple-Mail=_E46BB7BE-5C7F-47D1-ADCE-2633CEAEE2E4--

From dpreed@reed.com  Fri Jan  3 10:17:56 2014
Return-Path: <dpreed@reed.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 09D091ADFD4 for <aqm@ietfa.amsl.com>; Fri,  3 Jan 2014 10:17:56 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 0.8
X-Spam-Level: 
X-Spam-Status: No, score=0.8 tagged_above=-999 required=5 tests=[BAYES_50=0.8,  HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id DRb8C5GP94AR for <aqm@ietfa.amsl.com>; Fri,  3 Jan 2014 10:17:53 -0800 (PST)
Received: from smtp110.iad3a.emailsrvr.com (smtp110.iad3a.emailsrvr.com [173.203.187.110]) by ietfa.amsl.com (Postfix) with ESMTP id BD84A1ADFC8 for <aqm@ietf.org>; Fri,  3 Jan 2014 10:17:52 -0800 (PST)
Received: from localhost (localhost.localdomain [127.0.0.1]) by smtp6.relay.iad3a.emailsrvr.com (SMTP Server) with ESMTP id 111951A80D3; Fri,  3 Jan 2014 13:17:45 -0500 (EST)
X-Virus-Scanned: OK
Received: from app50.wa-webapps.iad3a (relay.iad3a.rsapps.net [172.27.255.110]) by smtp6.relay.iad3a.emailsrvr.com (SMTP Server) with ESMTP id E2E201A80C7; Fri,  3 Jan 2014 13:17:44 -0500 (EST)
Received: from reed.com (localhost.localdomain [127.0.0.1]) by app50.wa-webapps.iad3a (Postfix) with ESMTP id D2102380049; Fri,  3 Jan 2014 13:17:44 -0500 (EST)
Received: by apps.rackspace.com (Authenticated sender: dpreed@reed.com, from: dpreed@reed.com)  with HTTP; Fri, 3 Jan 2014 13:17:44 -0500 (EST)
Date: Fri, 3 Jan 2014 13:17:44 -0500 (EST)
From: dpreed@reed.com
To: "=?utf-8?Q?Fred_Baker_=28fred=29?=" <fred@cisco.com>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="----=_20140103131744000000_67191"
Importance: Normal
X-Priority: 3 (Normal)
X-Type: html
In-Reply-To: <533BE7A9-7804-4A74-BBFB-C75CCE212434@cisco.com>
References: <201312151857.rBFIuuea043478@gateway0.ipv6.occnc.com>  <533BE7A9-7804-4A74-BBFB-C75CCE212434@cisco.com>
Message-ID: <1388773064.857924560@apps.rackspace.com>
X-Mailer: webmail7.0
Cc: bloat <bloat@lists.bufferbloat.net>, "aqm@ietf.org" <aqm@ietf.org>, "=?utf-8?Q?=3Ccurtis=40ipv6.occnc.com=3E?=" <curtis@ipv6.occnc.com>
Subject: Re: [aqm] =?utf-8?q?=5Be2e=5D__What_is_a_good_burst=3F_--_AQM_evaluat?= =?utf-8?q?ion_guidelines?=
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 03 Jan 2014 18:17:56 -0000

------=_20140103131744000000_67191
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

=0AEnd-to-end queueing delay (aggregate of delays in all queues except for =
the queues in the endpoints themselves) should typically never (never means=
 <99.9% of any hour-long period)  exceed 200 msec. in the worst case, and i=
f at all possible never exceed 100 msec.   in networks capable of carrying =
more than 1 Mbit/sec to and from endpoints (I would call that high-bitrate =
nets, the stage up from "dialup" networks).=0A =0AThere are two reasons for=
 this:=0A =0A1) round-trip "RPC" response time for interactive applications=
 > 100 msec. become unreasonable.=0A =0A2) flow control at the source that =
stanches the entry of data into the network (which can be either switching =
media codecs or just pushing back on the application rate - whether it is d=
riven by the receiver or the sender, both of which are common) must respond=
 quickly, lest more packets be dumped into the network that sustain congest=
ion.=0A =0AFairness is a different axis, but I strongly suggest that there =
are other ways to achieve approximate fairness of any desired type without =
building up queues in routers.  It's perfectly reasonable to remember (in a=
ll the memory that *would otherwise have caused trouble by holding packets =
rather than discarding them*) the source/dest information and sizes of rece=
ntly processed (forwarded or discarded) packets.  This information takes le=
ss space than the packets themselves, of course!  It can even be further co=
mpressed by "coding or hashing" techniques.  Such live data about *recent b=
ehavior* is all you need for fairness in balancing signaling back to the so=
urce.=0A =0AIf all of the brainpower on this list cannot take that previous=
 paragraph and expand it to implement the solution I am talking about, I wo=
uld be happy (at my consulting rates, which are quite high) to write the co=
de for you.  But I have a day job that involves low-level scheduling and qu=
eueing work in a different domain of application.=0A =0ACan we please get r=
id of the nonsense that implies that the only information one can have at a=
 router/switch is the set of packets that are clogging its outbound queues?=
  Study some computer algorithms that provide memory of recent history.... =
 and please, please, please stop insisting that intra-network queues should=
 build up for any reason whatsoever other than instantaneous transient burs=
tiness of convergent traffic.  They should persist as briefly as possible, =
and not be sustained for some kind of "optimum" throughput that can be gain=
ed by reframing the problem.=0A =0A =0A =0A=0A=0AOn Thursday, January 2, 20=
14 1:31am, "Fred Baker (fred)" <fred@cisco.com> said:=0A=0A=0A=0A> =0A> On =
Dec 15, 2013, at 10:56 AM, Curtis Villamizar <curtis@ipv6.occnc.com>=0A> wr=
ote:=0A> =0A> > So briefly, my answer is: as a WG, I don't think we want to=
 go there.=0A> > If we do go there at all, then we should define "good AQM"=
 in terms of=0A> > acheving a "good" tradeoff between fairness, bulk transf=
er goodput,=0A> > and bounded delay.  IMHO sometimes vague is better.=0A> =
=0A> As you may have worked out from my previous comments in these threads,=
 I agree=0A> with you. I don't think this can be nailed down in a universal=
 sense. What can be=0A> described is the result in the network, in that del=
ays build up that persist, as=0A> opposed to coming and going, and as a res=
ult applications don't work as well as=0A> they might - and at that point, =
it is appropriate for the network to inform the=0A> transport.=0A>
------=_20140103131744000000_67191
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<font face=3D"arial" size=3D"2"><p style=3D"margin:0;padding:0;">End-to-end=
 queueing delay (aggregate of delays in all queues except for the queues in=
 the endpoints themselves) should typically never (never means &lt;99.9% of=
 any hour-long period) &nbsp;exceed 200 msec. in the worst case, and if at =
all possible never exceed 100 msec. &nbsp; in networks capable of carrying =
more than 1 Mbit/sec to and from endpoints (I would call that high-bitrate =
nets, the stage up from "dialup" networks).</p>=0A<p style=3D"margin:0;padd=
ing:0;">&nbsp;</p>=0A<p style=3D"margin:0;padding:0;">There are two reasons=
 for this:</p>=0A<p style=3D"margin:0;padding:0;">&nbsp;</p>=0A<p style=3D"=
margin:0;padding:0;">1) round-trip "RPC" response time for interactive appl=
ications &gt; 100 msec. become unreasonable.</p>=0A<p style=3D"margin:0;pad=
ding:0;">&nbsp;</p>=0A<p style=3D"margin:0;padding:0;">2) flow control at t=
he source that stanches the entry of data into the network (which can be ei=
ther switching media codecs or just pushing back on the application rate - =
whether it is driven by the receiver or the sender, both of which are commo=
n) must respond quickly, lest more packets be dumped into the network that =
sustain congestion.</p>=0A<p style=3D"margin:0;padding:0;">&nbsp;</p>=0A<p =
style=3D"margin:0;padding:0;">Fairness is a different axis, but I strongly =
suggest that there are other ways to achieve approximate fairness of any de=
sired type without building up queues in routers. &nbsp;It's perfectly reas=
onable to remember (in all the memory that *would otherwise have caused tro=
uble by holding packets rather than discarding them*) the source/dest infor=
mation and sizes of recently processed (forwarded or discarded) packets. &n=
bsp;This information takes less space than the packets themselves, of cours=
e! &nbsp;It can even be further compressed by "coding or hashing" technique=
s. &nbsp;Such live data about *recent behavior* is all you need for fairnes=
s in balancing signaling back to the source.</p>=0A<p style=3D"margin:0;pad=
ding:0;">&nbsp;</p>=0A<p style=3D"margin:0;padding:0;">If all of the brainp=
ower on this list cannot take that previous paragraph and expand it to impl=
ement the solution I am talking about, I would be happy (at my consulting r=
ates, which are quite high) to write the code for you. &nbsp;But I have a d=
ay job that involves low-level scheduling and queueing work in a different =
domain of application.</p>=0A<p style=3D"margin:0;padding:0;">&nbsp;</p>=0A=
<p style=3D"margin:0;padding:0;">Can we please get rid of the nonsense that=
 implies that the only information one can have at a router/switch is the s=
et of packets that are clogging its outbound queues? &nbsp;Study some compu=
ter algorithms that provide memory of recent history.... &nbsp;and please, =
please, please stop insisting that intra-network queues should build up for=
 any reason whatsoever other than instantaneous transient burstiness of con=
vergent traffic. &nbsp;They should persist as briefly as possible, and not =
be sustained for some kind of "optimum" throughput that can be gained by re=
framing the problem.</p>=0A<p style=3D"margin:0;padding:0;">&nbsp;</p>=0A<p=
 style=3D"margin:0;padding:0;">&nbsp;</p>=0A<p style=3D"margin:0;padding:0;=
">&nbsp;</p>=0A<!--WM_COMPOSE_SIGNATURE_START--><!--WM_COMPOSE_SIGNATURE_EN=
D-->=0A<p style=3D"margin:0;padding:0;"><br /><br />On Thursday, January 2,=
 2014 1:31am, "Fred Baker (fred)" &lt;fred@cisco.com&gt; said:<br /><br /><=
/p>=0A<div id=3D"SafeStyles1388771952">=0A<p style=3D"margin:0;padding:0;">=
&gt; <br />&gt; On Dec 15, 2013, at 10:56 AM, Curtis Villamizar &lt;curtis@=
ipv6.occnc.com&gt;<br />&gt; wrote:<br />&gt; <br />&gt; &gt; So briefly, m=
y answer is: as a WG, I don't think we want to go there.<br />&gt; &gt; If =
we do go there at all, then we should define "good AQM" in terms of<br />&g=
t; &gt; acheving a "good" tradeoff between fairness, bulk transfer goodput,=
<br />&gt; &gt; and bounded delay.  IMHO sometimes vague is better.<br />&g=
t; <br />&gt; As you may have worked out from my previous comments in these=
 threads, I agree<br />&gt; with you. I don't think this can be nailed down=
 in a universal sense. What can be<br />&gt; described is the result in the=
 network, in that delays build up that persist, as<br />&gt; opposed to com=
ing and going, and as a result applications don't work as well as<br />&gt;=
 they might - and at that point, it is appropriate for the network to infor=
m the<br />&gt; transport.<br />&gt;</p>=0A</div></font>
------=_20140103131744000000_67191--


From wes@mti-systems.com  Wed Jan  8 11:46:26 2014
Return-Path: <wes@mti-systems.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E968E1AE561 for <aqm@ietfa.amsl.com>; Wed,  8 Jan 2014 11:46:26 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Y6XIAyeu7y38 for <aqm@ietfa.amsl.com>; Wed,  8 Jan 2014 11:46:25 -0800 (PST)
Received: from atl4mhob14.myregisteredsite.com (atl4mhob14.myregisteredsite.com [209.17.115.52]) by ietfa.amsl.com (Postfix) with ESMTP id 23EAF1AE14F for <aqm@ietf.org>; Wed,  8 Jan 2014 11:46:25 -0800 (PST)
Received: from mailpod.hostingplatform.com ([10.30.71.210]) by atl4mhob14.myregisteredsite.com (8.14.4/8.14.4) with ESMTP id s08JkEFT025503 for <aqm@ietf.org>; Wed, 8 Jan 2014 14:46:14 -0500
Received: (qmail 1640 invoked by uid 0); 8 Jan 2014 19:46:14 -0000
X-TCPREMOTEIP: 107.45.8.204
X-Authenticated-UID: wes@mti-systems.com
Received: from unknown (HELO ?192.168.43.65?) (wes@mti-systems.com@107.45.8.204) by 0 with ESMTPA; 8 Jan 2014 19:46:12 -0000
Message-ID: <52CDAAF7.2060307@mti-systems.com>
Date: Wed, 08 Jan 2014 14:45:59 -0500
From: Wesley Eddy <wes@mti-systems.com>
Organization: MTI Systems
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:24.0) Gecko/20100101 Thunderbird/24.2.0
MIME-Version: 1.0
To: "aqm@ietf.org" <aqm@ietf.org>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Subject: [aqm] WG status
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 08 Jan 2014 19:46:27 -0000

Hi, as we've entered 2014 and have charter milestones that we're
working towards, Richard and I thought it would be good to start
periodically sending a "status report" to the WG mailing list so
that we can all keep up with what's going on, and focus our efforts
together on the things that need work.

Towards that goal, here is a snapshot of where we think the AQM
working group is at today, and what the next steps are that people
can contribute to:


- WG Milestones:
  - Submit AQM recommendations to IESG for publication, obsoleting RFC
2309 (Goal: January 2014)
    - draft-ietf-aqm-recommendation is accepted towards this milestone
    - http://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/
    - the draft needs to be updated per comments received, including
feedback on the recommendations from the Vancouver meeting
    - if the authors are comfortable, a WGLC might be made on the next
revision
    - we would like to hear from other authors of RFC 2309 on this
document, if anyone has contacts to them.

  - Submit AQM algorithm evaluation guidelines to IESG for publication
as Informational (Goal: July 2014)
    - We need an editor team to step forward and begin work on this;
there was initial work presented in Vancouver, but no draft available or
adopted by the working group yet.
    - It will be difficult to make this milestone, and will push other
milestones back, if this work isn't accelerated.
    - Please express interest to the chairs or on-list

  - Submit first algorithm specification to IESG for publication as
Proposed Standard (Goal: December 2014)
    - Since any Proposed Standard algorithm should be in line with the
recommendations and be passable versus the evaluation guidelines, this
milestone is hard to start on without significant progress on the
previous two.
    - Currently the only algorithm spec with a complete and active
individual-submission draft is PIE

- Other items:
  - draft-pan-aqm-pie is under active work as a proposed algorithm:
    http://tools.ietf.org/id/draft-pan-aqm-pie-00.txt
  - CoDel draft is expired; Dave Taht or others may revive it and/or
describe pairing with FQ/SFQ algorithms:
    http://tools.ietf.org/id/draft-nichols-tsvwg-codel-01.txt
  - Other algorithm specifications are welcome!
    - Though, we are not planning on adopting algorithms until
recommendations and evaluation guidelines are mostly stable


-- 
Wes Eddy
MTI Systems

From dave.taht@gmail.com  Wed Jan  8 12:06:16 2014
Return-Path: <dave.taht@gmail.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 090F31AE164 for <aqm@ietfa.amsl.com>; Wed,  8 Jan 2014 12:06:16 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id dEe9skch3hlR for <aqm@ietfa.amsl.com>; Wed,  8 Jan 2014 12:06:13 -0800 (PST)
Received: from mail-wg0-x233.google.com (mail-wg0-x233.google.com [IPv6:2a00:1450:400c:c00::233]) by ietfa.amsl.com (Postfix) with ESMTP id EA2AE1ADFA6 for <aqm@ietf.org>; Wed,  8 Jan 2014 12:06:12 -0800 (PST)
Received: by mail-wg0-f51.google.com with SMTP id z12so1708815wgg.18 for <aqm@ietf.org>; Wed, 08 Jan 2014 12:06:03 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=lSFxyJ15CJXp/4wKoKTt5tqb70LtvaKXvfEXrc+juME=; b=GBdzw9YWMmpXEhvC5QwXeBv14LGH8wG8nbpiUXmKrilIYLMPIbyNUueErlJ/Sw6rze N5fMzMSb7htmXmrrqfrcBJykOEpdh8DiABnnjKgSa3wA+A6wWBNpGm3O0FKPPG7o/3cg NZ/VO30mKG1uvH391lT+LZBwJxIGfcSccVsn1rm88CqPjdjVgM/HGYUuxU+op5oFjXJh I5wZ+P/MqnZ4i/lRaIMk6+g5g1a0HOi6vpR0y2lbbbKq9YDd4I5Zy4WtzgrKiy+MF8NY UPyuiJ2C5wcjRXEBu8lzpxlP0GItqaO81cjcX8/GuMM4WdV7BcOB0z4sm/e2taWnAgt8 AKrw==
MIME-Version: 1.0
X-Received: by 10.180.14.37 with SMTP id m5mr21669755wic.46.1389211563129; Wed, 08 Jan 2014 12:06:03 -0800 (PST)
Received: by 10.217.123.69 with HTTP; Wed, 8 Jan 2014 12:06:02 -0800 (PST)
Received: by 10.217.123.69 with HTTP; Wed, 8 Jan 2014 12:06:02 -0800 (PST)
In-Reply-To: <52CDAAF7.2060307@mti-systems.com>
References: <52CDAAF7.2060307@mti-systems.com>
Date: Wed, 8 Jan 2014 15:06:02 -0500
Message-ID: <CAA93jw5fvBAw-aayThod6egi_2JiZAc0E9Y4XfeXyqgYPZ2ncA@mail.gmail.com>
From: Dave Taht <dave.taht@gmail.com>
To: Wesley Eddy <wes@mti-systems.com>
Content-Type: multipart/alternative; boundary=f46d04138ca7bcff9904ef7b0594
Cc: Alaric Moore <alaricmeister@gmail.com>, aqm@ietf.org
Subject: Re: [aqm] WG status
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 08 Jan 2014 20:06:16 -0000

--f46d04138ca7bcff9904ef7b0594
Content-Type: text/plain; charset=ISO-8859-1

I have a doc shepard/researcher working with me to nail down cites from the
70s and 80s. First up however was finding direct web references to
everything cited in the existing  drafts.

Alaric found links to everything missing and modified the existing drafts
to suit. I don't know how the various authors would like the changes (?) -
diffs can be had from the bufferbloat-rfcs repository in github:
https://github.com/dtaht/bufferbloat-rfcs - there were only about 10 URLs
added.

Still struggling with xml2rfc and pandoc2rfc here...
 On Jan 8, 2014 2:46 PM, "Wesley Eddy" <wes@mti-systems.com> wrote:

> Hi, as we've entered 2014 and have charter milestones that we're
> working towards, Richard and I thought it would be good to start
> periodically sending a "status report" to the WG mailing list so
> that we can all keep up with what's going on, and focus our efforts
> together on the things that need work.
>
> Towards that goal, here is a snapshot of where we think the AQM
> working group is at today, and what the next steps are that people
> can contribute to:
>
>
> - WG Milestones:
>   - Submit AQM recommendations to IESG for publication, obsoleting RFC
> 2309 (Goal: January 2014)
>     - draft-ietf-aqm-recommendation is accepted towards this milestone
>     - http://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/
>     - the draft needs to be updated per comments received, including
> feedback on the recommendations from the Vancouver meeting
>     - if the authors are comfortable, a WGLC might be made on the next
> revision
>     - we would like to hear from other authors of RFC 2309 on this
> document, if anyone has contacts to them.
>
>   - Submit AQM algorithm evaluation guidelines to IESG for publication
> as Informational (Goal: July 2014)
>     - We need an editor team to step forward and begin work on this;
> there was initial work presented in Vancouver, but no draft available or
> adopted by the working group yet.
>     - It will be difficult to make this milestone, and will push other
> milestones back, if this work isn't accelerated.
>     - Please express interest to the chairs or on-list
>
>   - Submit first algorithm specification to IESG for publication as
> Proposed Standard (Goal: December 2014)
>     - Since any Proposed Standard algorithm should be in line with the
> recommendations and be passable versus the evaluation guidelines, this
> milestone is hard to start on without significant progress on the
> previous two.
>     - Currently the only algorithm spec with a complete and active
> individual-submission draft is PIE
>
> - Other items:
>   - draft-pan-aqm-pie is under active work as a proposed algorithm:
>     http://tools.ietf.org/id/draft-pan-aqm-pie-00.txt
>   - CoDel draft is expired; Dave Taht or others may revive it and/or
> describe pairing with FQ/SFQ algorithms:
>     http://tools.ietf.org/id/draft-nichols-tsvwg-codel-01.txt
>   - Other algorithm specifications are welcome!
>     - Though, we are not planning on adopting algorithms until
> recommendations and evaluation guidelines are mostly stable
>
>
> --
> Wes Eddy
> MTI Systems
> _______________________________________________
> aqm mailing list
> aqm@ietf.org
> https://www.ietf.org/mailman/listinfo/aqm
>

--f46d04138ca7bcff9904ef7b0594
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<p dir=3D"ltr">I have a doc shepard/researcher working with me to nail down=
 cites from the 70s and 80s. First up however was finding direct web refere=
nces to everything cited in the existing=A0 drafts.</p>
<p dir=3D"ltr">Alaric found links to everything missing and modified the ex=
isting drafts to suit. I don&#39;t know how the various authors would like =
the changes (?) - diffs can be had from the bufferbloat-rfcs repository in =
github:<br>

<a href=3D"https://github.com/dtaht/bufferbloat-rfcs">https://github.com/dt=
aht/bufferbloat-rfcs</a> - there were only about 10 URLs added.</p>
<p dir=3D"ltr">Still struggling with xml2rfc and pandoc2rfc here...<br>
</p>
<div class=3D"gmail_quote">On Jan 8, 2014 2:46 PM, &quot;Wesley Eddy&quot; =
&lt;<a href=3D"mailto:wes@mti-systems.com">wes@mti-systems.com</a>&gt; wrot=
e:<br type=3D"attribution"><blockquote class=3D"gmail_quote" style=3D"margi=
n:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hi, as we&#39;ve entered 2014 and have charter milestones that we&#39;re<br=
>
working towards, Richard and I thought it would be good to start<br>
periodically sending a &quot;status report&quot; to the WG mailing list so<=
br>
that we can all keep up with what&#39;s going on, and focus our efforts<br>
together on the things that need work.<br>
<br>
Towards that goal, here is a snapshot of where we think the AQM<br>
working group is at today, and what the next steps are that people<br>
can contribute to:<br>
<br>
<br>
- WG Milestones:<br>
=A0 - Submit AQM recommendations to IESG for publication, obsoleting RFC<br=
>
2309 (Goal: January 2014)<br>
=A0 =A0 - draft-ietf-aqm-recommendation is accepted towards this milestone<=
br>
=A0 =A0 - <a href=3D"http://datatracker.ietf.org/doc/draft-ietf-aqm-recomme=
ndation/" target=3D"_blank">http://datatracker.ietf.org/doc/draft-ietf-aqm-=
recommendation/</a><br>
=A0 =A0 - the draft needs to be updated per comments received, including<br=
>
feedback on the recommendations from the Vancouver meeting<br>
=A0 =A0 - if the authors are comfortable, a WGLC might be made on the next<=
br>
revision<br>
=A0 =A0 - we would like to hear from other authors of RFC 2309 on this<br>
document, if anyone has contacts to them.<br>
<br>
=A0 - Submit AQM algorithm evaluation guidelines to IESG for publication<br=
>
as Informational (Goal: July 2014)<br>
=A0 =A0 - We need an editor team to step forward and begin work on this;<br=
>
there was initial work presented in Vancouver, but no draft available or<br=
>
adopted by the working group yet.<br>
=A0 =A0 - It will be difficult to make this milestone, and will push other<=
br>
milestones back, if this work isn&#39;t accelerated.<br>
=A0 =A0 - Please express interest to the chairs or on-list<br>
<br>
=A0 - Submit first algorithm specification to IESG for publication as<br>
Proposed Standard (Goal: December 2014)<br>
=A0 =A0 - Since any Proposed Standard algorithm should be in line with the<=
br>
recommendations and be passable versus the evaluation guidelines, this<br>
milestone is hard to start on without significant progress on the<br>
previous two.<br>
=A0 =A0 - Currently the only algorithm spec with a complete and active<br>
individual-submission draft is PIE<br>
<br>
- Other items:<br>
=A0 - draft-pan-aqm-pie is under active work as a proposed algorithm:<br>
=A0 =A0 <a href=3D"http://tools.ietf.org/id/draft-pan-aqm-pie-00.txt" targe=
t=3D"_blank">http://tools.ietf.org/id/draft-pan-aqm-pie-00.txt</a><br>
=A0 - CoDel draft is expired; Dave Taht or others may revive it and/or<br>
describe pairing with FQ/SFQ algorithms:<br>
=A0 =A0 <a href=3D"http://tools.ietf.org/id/draft-nichols-tsvwg-codel-01.tx=
t" target=3D"_blank">http://tools.ietf.org/id/draft-nichols-tsvwg-codel-01.=
txt</a><br>
=A0 - Other algorithm specifications are welcome!<br>
=A0 =A0 - Though, we are not planning on adopting algorithms until<br>
recommendations and evaluation guidelines are mostly stable<br>
<br>
<br>
--<br>
Wes Eddy<br>
MTI Systems<br>
_______________________________________________<br>
aqm mailing list<br>
<a href=3D"mailto:aqm@ietf.org">aqm@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/aqm" target=3D"_blank">htt=
ps://www.ietf.org/mailman/listinfo/aqm</a><br>
</blockquote></div>

--f46d04138ca7bcff9904ef7b0594--

From preethi.cis@gmail.com  Wed Jan  8 12:53:49 2014
Return-Path: <preethi.cis@gmail.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E32E91AE1C0 for <aqm@ietfa.amsl.com>; Wed,  8 Jan 2014 12:53:49 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gsfBn4Rye_rL for <aqm@ietfa.amsl.com>; Wed,  8 Jan 2014 12:53:48 -0800 (PST)
Received: from mail-pd0-x22c.google.com (mail-pd0-x22c.google.com [IPv6:2607:f8b0:400e:c02::22c]) by ietfa.amsl.com (Postfix) with ESMTP id A97F31AE19F for <aqm@ietf.org>; Wed,  8 Jan 2014 12:53:48 -0800 (PST)
Received: by mail-pd0-f172.google.com with SMTP id g10so2263236pdj.3 for <aqm@ietf.org>; Wed, 08 Jan 2014 12:53:39 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=user-agent:date:subject:from:to:message-id:thread-topic:in-reply-to :mime-version:content-type:content-transfer-encoding; bh=v9Cb1HoAdgH+aFPgm1Y2fQYsmmnAPrOJZKN2Ac3imvA=; b=g3lgMxc0f5b3T16b8QyCuGWTKcdbRTnJrEu4SlzAbcjNIYte6XzofzBivlCCisdpn0 SFf2nuskjwUDbNqFhzgpNrymlxOGsP3mFuMvozCPPVQWk2ezIyPEEQWjKjrhYspqCmNv 48C6IxzmYmre7PB9mM1vSNjnivSeE4NSUmxSdMngU13JE56KGM42D7kOp5q02XLrpzox Yq4bQUyzDX/wzmdaLdh/QOLA8BFPTk7RJx1hH+jVqda4cd95frMneA/nxdior8DMvEUr Eg5rHOEGORoqO39e1JPDZij+MC1N5sUz2fRuUUxCRo8jQQS2kVI49dJcmZAgL9f+tfkP yQqQ==
X-Received: by 10.68.129.201 with SMTP id ny9mr145736649pbb.70.1389214419458;  Wed, 08 Jan 2014 12:53:39 -0800 (PST)
Received: from [192.168.1.104] (c-76-103-130-90.hsd1.ca.comcast.net. [76.103.130.90]) by mx.google.com with ESMTPSA id gg10sm4569810pbc.46.2014.01.08.12.53.36 for <multiple recipients> (version=TLSv1 cipher=RC4-SHA bits=128/128); Wed, 08 Jan 2014 12:53:39 -0800 (PST)
User-Agent: Microsoft-MacOutlook/14.2.1.120420
Date: Wed, 08 Jan 2014 12:53:25 -0800
From: Preethi Natarajan <preethi.cis@gmail.com>
To: Wesley Eddy <wes@mti-systems.com>, "aqm@ietf.org" <aqm@ietf.org>
Message-ID: <CEF2FAA2.4E02E%prenatar@cisco.com>
Thread-Topic: [aqm] WG status
In-Reply-To: <52CDAAF7.2060307@mti-systems.com>
Mime-version: 1.0
Content-type: text/plain; charset="US-ASCII"
Content-transfer-encoding: 7bit
Subject: Re: [aqm] WG status
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 08 Jan 2014 20:53:50 -0000

On 1/8/14 11:45 AM, "Wesley Eddy" <wes@mti-systems.com> wrote:

>
>  - Submit AQM algorithm evaluation guidelines to IESG for publication
>as Informational (Goal: July 2014)
>    - We need an editor team to step forward and begin work on this;
>there was initial work presented in Vancouver, but no draft available or
>adopted by the working group yet.
>    - It will be difficult to make this milestone, and will push other
>milestones back, if this work isn't accelerated.
>    - Please express interest to the chairs or on-list


Interested in contributing towards the evaluation guidelines draft --
write, edit, whatever else is needed to meet deadline.

Thanks,
Preethi (on behalf of PIE team)




From naeem.khademi@gmail.com  Wed Jan  8 13:03:00 2014
Return-Path: <naeem.khademi@gmail.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 505631AE5CC for <aqm@ietfa.amsl.com>; Wed,  8 Jan 2014 13:03:00 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id W5Sqo3vV-CS5 for <aqm@ietfa.amsl.com>; Wed,  8 Jan 2014 13:02:58 -0800 (PST)
Received: from mail-ve0-x230.google.com (mail-ve0-x230.google.com [IPv6:2607:f8b0:400c:c01::230]) by ietfa.amsl.com (Postfix) with ESMTP id 0CA041AE5C8 for <aqm@ietf.org>; Wed,  8 Jan 2014 13:02:57 -0800 (PST)
Received: by mail-ve0-f176.google.com with SMTP id oz11so1732212veb.35 for <aqm@ietf.org>; Wed, 08 Jan 2014 13:02:48 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=RytbbduQKZ2o4CK4uHA8Haq3xQb77Zu/1+FaRFp7Nqs=; b=SBhHliPNfCpwaKTnaMpks4sqUiNEXfTlkakrK4vyqld9SaxdTCEDa8vohPnM1Z2af8 uh3hpDZovAZBH9FFcAScugCS9IY/Pu7VDy6Llu7hUo7okbWW1vQFOSFJuabpMj0glpGa wdAZ3LbudniRzsZOG87Y5txGWD51u2xaX3NOWryqWuzVldF+zYziSEESrUw9u1rAh5JM zvvfZUJsBGi6MT7vxKt4Vehbhu2mh4EnmLye8sey92PmIvE9hLW76gCRs91yQ0zNO44F uG24gDljCNThMpCv/YTZ1AwX4IFRbOzAqVgr1RWFaHxKxxVcjCyjHJPE3uLIARl5KMoq 62gg==
MIME-Version: 1.0
X-Received: by 10.58.208.130 with SMTP id me2mr85250267vec.13.1389214968529; Wed, 08 Jan 2014 13:02:48 -0800 (PST)
Received: by 10.58.90.197 with HTTP; Wed, 8 Jan 2014 13:02:48 -0800 (PST)
In-Reply-To: <52CDAAF7.2060307@mti-systems.com>
References: <52CDAAF7.2060307@mti-systems.com>
Date: Wed, 8 Jan 2014 22:02:48 +0100
Message-ID: <CAEjQQ5U5sJhaoM64V0MTjhKNaiqf8Wu2OA=OqOAYQhEPRLTepw@mail.gmail.com>
From: Naeem Khademi <naeem.khademi@gmail.com>
To: Wesley Eddy <wes@mti-systems.com>
Content-Type: multipart/alternative; boundary=047d7bdc192cb74c9204ef7bd0ea
Cc: "aqm@ietf.org" <aqm@ietf.org>
Subject: Re: [aqm] WG status
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 08 Jan 2014 21:03:00 -0000

--047d7bdc192cb74c9204ef7bd0ea
Content-Type: text/plain; charset=ISO-8859-1

I'm interested in contributing to the evaluation guidelines draft as writer
as expressed before, although I won't be able to actively work on this
earlier than the second week of Feb'14.

Regards,
Naeem




On Wed, Jan 8, 2014 at 8:45 PM, Wesley Eddy <wes@mti-systems.com> wrote:

> Hi, as we've entered 2014 and have charter milestones that we're
> working towards, Richard and I thought it would be good to start
> periodically sending a "status report" to the WG mailing list so
> that we can all keep up with what's going on, and focus our efforts
> together on the things that need work.
>
> Towards that goal, here is a snapshot of where we think the AQM
> working group is at today, and what the next steps are that people
> can contribute to:
>
>
> - WG Milestones:
>   - Submit AQM recommendations to IESG for publication, obsoleting RFC
> 2309 (Goal: January 2014)
>     - draft-ietf-aqm-recommendation is accepted towards this milestone
>     - http://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/
>     - the draft needs to be updated per comments received, including
> feedback on the recommendations from the Vancouver meeting
>     - if the authors are comfortable, a WGLC might be made on the next
> revision
>     - we would like to hear from other authors of RFC 2309 on this
> document, if anyone has contacts to them.
>
>   - Submit AQM algorithm evaluation guidelines to IESG for publication
> as Informational (Goal: July 2014)
>     - We need an editor team to step forward and begin work on this;
> there was initial work presented in Vancouver, but no draft available or
> adopted by the working group yet.
>     - It will be difficult to make this milestone, and will push other
> milestones back, if this work isn't accelerated.
>     - Please express interest to the chairs or on-list
>
>   - Submit first algorithm specification to IESG for publication as
> Proposed Standard (Goal: December 2014)
>     - Since any Proposed Standard algorithm should be in line with the
> recommendations and be passable versus the evaluation guidelines, this
> milestone is hard to start on without significant progress on the
> previous two.
>     - Currently the only algorithm spec with a complete and active
> individual-submission draft is PIE
>
> - Other items:
>   - draft-pan-aqm-pie is under active work as a proposed algorithm:
>     http://tools.ietf.org/id/draft-pan-aqm-pie-00.txt
>   - CoDel draft is expired; Dave Taht or others may revive it and/or
> describe pairing with FQ/SFQ algorithms:
>     http://tools.ietf.org/id/draft-nichols-tsvwg-codel-01.txt
>   - Other algorithm specifications are welcome!
>     - Though, we are not planning on adopting algorithms until
> recommendations and evaluation guidelines are mostly stable
>
>
> --
> Wes Eddy
> MTI Systems
> _______________________________________________
> aqm mailing list
> aqm@ietf.org
> https://www.ietf.org/mailman/listinfo/aqm
>

--047d7bdc192cb74c9204ef7bd0ea
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">I&#39;m i<span style=3D"font-family:arial,sans-serif;font-=
size:13px">nterested in contributing to the evaluation guidelines draft as =
writer as expressed before, although I won&#39;t be able to actively work o=
n this earlier than the second week of Feb&#39;14.=A0</span><div>
<span style=3D"font-family:arial,sans-serif;font-size:13px"><br></span></di=
v><div style><span style=3D"font-family:arial,sans-serif;font-size:13px">Re=
gards,</span></div><div><span style=3D"font-family:arial,sans-serif;font-si=
ze:13px">Naeem=A0</span><div>
<span style=3D"font-family:arial,sans-serif;font-size:13px"><br></span></di=
v><div><span style=3D"font-family:arial,sans-serif;font-size:13px">=A0</spa=
n></div></div></div><div class=3D"gmail_extra"><br><br><div class=3D"gmail_=
quote">
On Wed, Jan 8, 2014 at 8:45 PM, Wesley Eddy <span dir=3D"ltr">&lt;<a href=
=3D"mailto:wes@mti-systems.com" target=3D"_blank">wes@mti-systems.com</a>&g=
t;</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0=
 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hi, as we&#39;ve entered 2014 and have charter milestones that we&#39;re<br=
>
working towards, Richard and I thought it would be good to start<br>
periodically sending a &quot;status report&quot; to the WG mailing list so<=
br>
that we can all keep up with what&#39;s going on, and focus our efforts<br>
together on the things that need work.<br>
<br>
Towards that goal, here is a snapshot of where we think the AQM<br>
working group is at today, and what the next steps are that people<br>
can contribute to:<br>
<br>
<br>
- WG Milestones:<br>
=A0 - Submit AQM recommendations to IESG for publication, obsoleting RFC<br=
>
2309 (Goal: January 2014)<br>
=A0 =A0 - draft-ietf-aqm-recommendation is accepted towards this milestone<=
br>
=A0 =A0 - <a href=3D"http://datatracker.ietf.org/doc/draft-ietf-aqm-recomme=
ndation/" target=3D"_blank">http://datatracker.ietf.org/doc/draft-ietf-aqm-=
recommendation/</a><br>
=A0 =A0 - the draft needs to be updated per comments received, including<br=
>
feedback on the recommendations from the Vancouver meeting<br>
=A0 =A0 - if the authors are comfortable, a WGLC might be made on the next<=
br>
revision<br>
=A0 =A0 - we would like to hear from other authors of RFC 2309 on this<br>
document, if anyone has contacts to them.<br>
<br>
=A0 - Submit AQM algorithm evaluation guidelines to IESG for publication<br=
>
as Informational (Goal: July 2014)<br>
=A0 =A0 - We need an editor team to step forward and begin work on this;<br=
>
there was initial work presented in Vancouver, but no draft available or<br=
>
adopted by the working group yet.<br>
=A0 =A0 - It will be difficult to make this milestone, and will push other<=
br>
milestones back, if this work isn&#39;t accelerated.<br>
=A0 =A0 - Please express interest to the chairs or on-list<br>
<br>
=A0 - Submit first algorithm specification to IESG for publication as<br>
Proposed Standard (Goal: December 2014)<br>
=A0 =A0 - Since any Proposed Standard algorithm should be in line with the<=
br>
recommendations and be passable versus the evaluation guidelines, this<br>
milestone is hard to start on without significant progress on the<br>
previous two.<br>
=A0 =A0 - Currently the only algorithm spec with a complete and active<br>
individual-submission draft is PIE<br>
<br>
- Other items:<br>
=A0 - draft-pan-aqm-pie is under active work as a proposed algorithm:<br>
=A0 =A0 <a href=3D"http://tools.ietf.org/id/draft-pan-aqm-pie-00.txt" targe=
t=3D"_blank">http://tools.ietf.org/id/draft-pan-aqm-pie-00.txt</a><br>
=A0 - CoDel draft is expired; Dave Taht or others may revive it and/or<br>
describe pairing with FQ/SFQ algorithms:<br>
=A0 =A0 <a href=3D"http://tools.ietf.org/id/draft-nichols-tsvwg-codel-01.tx=
t" target=3D"_blank">http://tools.ietf.org/id/draft-nichols-tsvwg-codel-01.=
txt</a><br>
=A0 - Other algorithm specifications are welcome!<br>
=A0 =A0 - Though, we are not planning on adopting algorithms until<br>
recommendations and evaluation guidelines are mostly stable<br>
<span class=3D"HOEnZb"><font color=3D"#888888"><br>
<br>
--<br>
Wes Eddy<br>
MTI Systems<br>
_______________________________________________<br>
aqm mailing list<br>
<a href=3D"mailto:aqm@ietf.org">aqm@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/aqm" target=3D"_blank">htt=
ps://www.ietf.org/mailman/listinfo/aqm</a><br>
</font></span></blockquote></div><br></div>

--047d7bdc192cb74c9204ef7bd0ea--

From davec-b@rogers.com  Wed Jan  8 13:20:47 2014
Return-Path: <davec-b@rogers.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C18271A1F1B for <aqm@ietfa.amsl.com>; Wed,  8 Jan 2014 13:20:47 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 03m9Ig0iAcFx for <aqm@ietfa.amsl.com>; Wed,  8 Jan 2014 13:20:44 -0800 (PST)
Received: from nm7-vm9.access.bullet.mail.gq1.yahoo.com (nm7-vm9.access.bullet.mail.gq1.yahoo.com [216.39.63.245]) by ietfa.amsl.com (Postfix) with ESMTP id C59A21A1F06 for <aqm@ietf.org>; Wed,  8 Jan 2014 13:20:44 -0800 (PST)
Received: from [216.39.60.173] by nm7.access.bullet.mail.gq1.yahoo.com with NNFMP; 08 Jan 2014 21:20:35 -0000
Received: from [67.195.22.113] by tm9.access.bullet.mail.gq1.yahoo.com with NNFMP; 08 Jan 2014 21:20:35 -0000
Received: from [127.0.0.1] by smtp115.sbc.mail.gq1.yahoo.com with NNFMP; 08 Jan 2014 21:20:35 -0000
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rogers.com; s=s1024; t=1389216035; bh=QZ0ktltcAS9H20nUe99pg/bjF3135gm34aZIK38Hlyw=; h=X-Yahoo-Newman-Id:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:X-Rocket-Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:Subject:References:In-Reply-To:X-Enigmail-Version:Content-Type; b=FcUxoYpc7WU2zHos6cADtLNTVBXRJeGi/4TnhXnNRCWfypvP7Sc5gY/EGHvtD8I+YYmgiSyeJzfwbzPZAH8XoPmPFtKzigvJfKIWyR5U0P0+ceh5RQXSTPhm3UOnmRUReHp9yNECsyxtT6vogwzxmqtjzi0gRCfnzrQbni3fYqg=
X-Yahoo-Newman-Id: 327654.78416.bm@smtp115.sbc.mail.gq1.yahoo.com
X-Yahoo-Newman-Property: ymail-3
X-YMail-OSG: YuECExsVM1lH82AtX6DnhU9QRyAWFeJoalqPXfFm0LMr9qc 29rHxenzgeYWe2H1pn9Aa4ACZpDc6ibQ4_sr4jmC9pngvvCJOw.D5LTWdRFS hFOxF0p2pvH0vhWGNC6R0JOJK62PJDXP0UREEeWaimLSgdJ5RNk9lQ6kZKKV f9l7jrdxiCTJgNd2Hn_.X1X1yWO9ZZm8OkLkfX5iWUSUDo2V5BLQsq1YR9vT bcS6swnqYsEtNM_tpbrL1kmhTPWpXUlKobRhD3fgF6SbMv8RYU9u_eYLGtIl YKsVynZRk_2sZrZXsY9l48NNuE5yDoE0nFL0A8ZM.qXkEc_Dfb4xV0GpPkH5 ROueKQPdwoocWdfzq1sE4MXB9XS1vBEv45Rp2E_ywfnTHdAQvAWHOghgK6wo e.O1pHhSLZvuDFOekt8zDiEl1H7ucB7UCMMi3wUIvJOOASFyMTfwaaj0o_HB M96PXaL22gyZWgz67zMYKFaiFUidMP7U3HDEBsUM7gfcdoT9fnLkxx.FjMbv M8123S8h4suMYWfCRNQOo6Klmo8ZUtYHCfSS6ydkgVLg2HmEUnDfBSvTkAuz Bg8hhRyccqV5hPyY63bs5uM9u_MYq2akUOzUOLyUGjTIqeJNQp2ZRQ8m1Xlc _TOAwTvEBQfTOhnPDQSvDXmw7Iuv.
X-Yahoo-SMTP: sltvjZWswBCRD.ElTuB1l9j6s9wRYPpuyTNWOE5oEg--
X-Rocket-Received: from [10.111.100.153] (davec-b@74.213.188.46 with plain [67.195.22.104]) by smtp115.sbc.mail.gq1.yahoo.com with SMTP; 08 Jan 2014 13:20:35 -0800 PST
Message-ID: <52CDC121.3090707@rogers.com>
Date: Wed, 08 Jan 2014 16:20:33 -0500
From: David Collier-Brown <davec-b@rogers.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7
MIME-Version: 1.0
To: aqm@ietf.org
References: <52CDAAF7.2060307@mti-systems.com> <CAA93jw5fvBAw-aayThod6egi_2JiZAc0E9Y4XfeXyqgYPZ2ncA@mail.gmail.com>
In-Reply-To: <CAA93jw5fvBAw-aayThod6egi_2JiZAc0E9Y4XfeXyqgYPZ2ncA@mail.gmail.com>
X-Enigmail-Version: 1.6
Content-Type: multipart/alternative; boundary="------------010101050307020608040709"
Subject: Re: [aqm] WG status
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: davecb@spamcop.net
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 08 Jan 2014 21:20:48 -0000

This is a multi-part message in MIME format.
--------------010101050307020608040709
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

I'll have a look and see if I can contribute.  --dave


On 01/08/2014 03:06 PM, Dave Taht wrote:
>
> I have a doc shepard/researcher working with me to nail down cites
> from the 70s and 80s. First up however was finding direct web
> references to everything cited in the existing  drafts.
>
> Alaric found links to everything missing and modified the existing
> drafts to suit. I don't know how the various authors would like the
> changes (?) - diffs can be had from the bufferbloat-rfcs repository in
> github:
> https://github.com/dtaht/bufferbloat-rfcs - there were only about 10
> URLs added.
>
> Still struggling with xml2rfc and pandoc2rfc here...
>
> On Jan 8, 2014 2:46 PM, "Wesley Eddy" <wes@mti-systems.com
> <mailto:wes@mti-systems.com>> wrote:
>
>     Hi, as we've entered 2014 and have charter milestones that we're
>     working towards, Richard and I thought it would be good to start
>     periodically sending a "status report" to the WG mailing list so
>     that we can all keep up with what's going on, and focus our efforts
>     together on the things that need work.
>
>     Towards that goal, here is a snapshot of where we think the AQM
>     working group is at today, and what the next steps are that people
>     can contribute to:
>
>
>     - WG Milestones:
>       - Submit AQM recommendations to IESG for publication, obsoleting RFC
>     2309 (Goal: January 2014)
>         - draft-ietf-aqm-recommendation is accepted towards this milestone
>         - http://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/
>         - the draft needs to be updated per comments received, including
>     feedback on the recommendations from the Vancouver meeting
>         - if the authors are comfortable, a WGLC might be made on the next
>     revision
>         - we would like to hear from other authors of RFC 2309 on this
>     document, if anyone has contacts to them.
>
>       - Submit AQM algorithm evaluation guidelines to IESG for publication
>     as Informational (Goal: July 2014)
>         - We need an editor team to step forward and begin work on this;
>     there was initial work presented in Vancouver, but no draft
>     available or
>     adopted by the working group yet.
>         - It will be difficult to make this milestone, and will push other
>     milestones back, if this work isn't accelerated.
>         - Please express interest to the chairs or on-list
>
>       - Submit first algorithm specification to IESG for publication as
>     Proposed Standard (Goal: December 2014)
>         - Since any Proposed Standard algorithm should be in line with the
>     recommendations and be passable versus the evaluation guidelines, this
>     milestone is hard to start on without significant progress on the
>     previous two.
>         - Currently the only algorithm spec with a complete and active
>     individual-submission draft is PIE
>
>     - Other items:
>       - draft-pan-aqm-pie is under active work as a proposed algorithm:
>         http://tools.ietf.org/id/draft-pan-aqm-pie-00.txt
>       - CoDel draft is expired; Dave Taht or others may revive it and/or
>     describe pairing with FQ/SFQ algorithms:
>         http://tools.ietf.org/id/draft-nichols-tsvwg-codel-01.txt
>       - Other algorithm specifications are welcome!
>         - Though, we are not planning on adopting algorithms until
>     recommendations and evaluation guidelines are mostly stable
>
>
>     --
>     Wes Eddy
>     MTI Systems
>     _______________________________________________
>     aqm mailing list
>     aqm@ietf.org <mailto:aqm@ietf.org>
>     https://www.ietf.org/mailman/listinfo/aqm
>
>
>
> _______________________________________________
> aqm mailing list
> aqm@ietf.org
> https://www.ietf.org/mailman/listinfo/aqm


-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
davecb@spamcop.net           |                      -- Mark Twain
(416) 223-8968


--------------010101050307020608040709
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">I'll have a look and see if I can
      contribute.&nbsp; --dave<br>
      <br>
      <br>
      On 01/08/2014 03:06 PM, Dave Taht wrote:<br>
    </div>
    <blockquote
cite="mid:CAA93jw5fvBAw-aayThod6egi_2JiZAc0E9Y4XfeXyqgYPZ2ncA@mail.gmail.com"
      type="cite">
      <p dir="ltr">I have a doc shepard/researcher working with me to
        nail down cites from the 70s and 80s. First up however was
        finding direct web references to everything cited in the
        existing&nbsp; drafts.</p>
      <p dir="ltr">Alaric found links to everything missing and modified
        the existing drafts to suit. I don't know how the various
        authors would like the changes (?) - diffs can be had from the
        bufferbloat-rfcs repository in github:<br>
        <a moz-do-not-send="true"
          href="https://github.com/dtaht/bufferbloat-rfcs">https://github.com/dtaht/bufferbloat-rfcs</a>
        - there were only about 10 URLs added.</p>
      <p dir="ltr">Still struggling with xml2rfc and pandoc2rfc here...<br>
      </p>
      <div class="gmail_quote">On Jan 8, 2014 2:46 PM, "Wesley Eddy"
        &lt;<a moz-do-not-send="true" href="mailto:wes@mti-systems.com">wes@mti-systems.com</a>&gt;
        wrote:<br type="attribution">
        <blockquote class="gmail_quote" style="margin:0 0 0
          .8ex;border-left:1px #ccc solid;padding-left:1ex">
          Hi, as we've entered 2014 and have charter milestones that
          we're<br>
          working towards, Richard and I thought it would be good to
          start<br>
          periodically sending a "status report" to the WG mailing list
          so<br>
          that we can all keep up with what's going on, and focus our
          efforts<br>
          together on the things that need work.<br>
          <br>
          Towards that goal, here is a snapshot of where we think the
          AQM<br>
          working group is at today, and what the next steps are that
          people<br>
          can contribute to:<br>
          <br>
          <br>
          - WG Milestones:<br>
          &nbsp; - Submit AQM recommendations to IESG for publication,
          obsoleting RFC<br>
          2309 (Goal: January 2014)<br>
          &nbsp; &nbsp; - draft-ietf-aqm-recommendation is accepted towards this
          milestone<br>
          &nbsp; &nbsp; - <a moz-do-not-send="true"
            href="http://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/"
            target="_blank">http://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/</a><br>
          &nbsp; &nbsp; - the draft needs to be updated per comments received,
          including<br>
          feedback on the recommendations from the Vancouver meeting<br>
          &nbsp; &nbsp; - if the authors are comfortable, a WGLC might be made on
          the next<br>
          revision<br>
          &nbsp; &nbsp; - we would like to hear from other authors of RFC 2309 on
          this<br>
          document, if anyone has contacts to them.<br>
          <br>
          &nbsp; - Submit AQM algorithm evaluation guidelines to IESG for
          publication<br>
          as Informational (Goal: July 2014)<br>
          &nbsp; &nbsp; - We need an editor team to step forward and begin work on
          this;<br>
          there was initial work presented in Vancouver, but no draft
          available or<br>
          adopted by the working group yet.<br>
          &nbsp; &nbsp; - It will be difficult to make this milestone, and will
          push other<br>
          milestones back, if this work isn't accelerated.<br>
          &nbsp; &nbsp; - Please express interest to the chairs or on-list<br>
          <br>
          &nbsp; - Submit first algorithm specification to IESG for
          publication as<br>
          Proposed Standard (Goal: December 2014)<br>
          &nbsp; &nbsp; - Since any Proposed Standard algorithm should be in line
          with the<br>
          recommendations and be passable versus the evaluation
          guidelines, this<br>
          milestone is hard to start on without significant progress on
          the<br>
          previous two.<br>
          &nbsp; &nbsp; - Currently the only algorithm spec with a complete and
          active<br>
          individual-submission draft is PIE<br>
          <br>
          - Other items:<br>
          &nbsp; - draft-pan-aqm-pie is under active work as a proposed
          algorithm:<br>
          &nbsp; &nbsp; <a moz-do-not-send="true"
            href="http://tools.ietf.org/id/draft-pan-aqm-pie-00.txt"
            target="_blank">http://tools.ietf.org/id/draft-pan-aqm-pie-00.txt</a><br>
          &nbsp; - CoDel draft is expired; Dave Taht or others may revive it
          and/or<br>
          describe pairing with FQ/SFQ algorithms:<br>
          &nbsp; &nbsp; <a moz-do-not-send="true"
            href="http://tools.ietf.org/id/draft-nichols-tsvwg-codel-01.txt"
            target="_blank">http://tools.ietf.org/id/draft-nichols-tsvwg-codel-01.txt</a><br>
          &nbsp; - Other algorithm specifications are welcome!<br>
          &nbsp; &nbsp; - Though, we are not planning on adopting algorithms until<br>
          recommendations and evaluation guidelines are mostly stable<br>
          <br>
          <br>
          --<br>
          Wes Eddy<br>
          MTI Systems<br>
          _______________________________________________<br>
          aqm mailing list<br>
          <a moz-do-not-send="true" href="mailto:aqm@ietf.org">aqm@ietf.org</a><br>
          <a moz-do-not-send="true"
            href="https://www.ietf.org/mailman/listinfo/aqm"
            target="_blank">https://www.ietf.org/mailman/listinfo/aqm</a><br>
        </blockquote>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
aqm mailing list
<a class="moz-txt-link-abbreviated" href="mailto:aqm@ietf.org">aqm@ietf.org</a>
<a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/aqm">https://www.ietf.org/mailman/listinfo/aqm</a>
</pre>
    </blockquote>
    <br>
    <br>
    <pre class="moz-signature" cols="72">-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
<a class="moz-txt-link-abbreviated" href="mailto:davecb@spamcop.net">davecb@spamcop.net</a>           |                      -- Mark Twain
(416) 223-8968
</pre>
  </body>
</html>

--------------010101050307020608040709--

From rs@netapp.com  Fri Jan 10 06:20:04 2014
Return-Path: <rs@netapp.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 017931AE064 for <aqm@ietfa.amsl.com>; Fri, 10 Jan 2014 06:20:04 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.44
X-Spam-Level: 
X-Spam-Status: No, score=-2.44 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, RP_MATCHES_RCVD=-0.538, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EQYQeLsoTJoy for <aqm@ietfa.amsl.com>; Fri, 10 Jan 2014 06:20:02 -0800 (PST)
Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id 3A3911AE062 for <aqm@ietf.org>; Fri, 10 Jan 2014 06:20:02 -0800 (PST)
X-IronPort-AV: E=Sophos;i="4.95,638,1384329600"; d="scan'208";a="95279574"
Received: from vmwexceht05-prd.hq.netapp.com ([10.106.77.35]) by mx11-out.netapp.com with ESMTP; 10 Jan 2014 06:19:51 -0800
Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.147]) by vmwexceht05-prd.hq.netapp.com ([10.106.77.35]) with mapi id 14.03.0123.003; Fri, 10 Jan 2014 06:19:51 -0800
From: "Scheffenegger, Richard" <rs@netapp.com>
To: Preethi Natarajan <preethi.cis@gmail.com>, "aqm@ietf.org" <aqm@ietf.org>,  Naeem Khademi <naeem.khademi@gmail.com>, David Collier-Brown <davec-b@rogers.com>
Thread-Topic: [aqm] WG status
Thread-Index: AQHPDKpjfDc4O/iDH0qQCt7bybJXrpp71CCAgAIu6ZA=
Date: Fri, 10 Jan 2014 14:19:50 +0000
Message-ID: <012C3117EDDB3C4781FD802A8C27DD4F25F8BC5D@SACEXCMBX02-PRD.hq.netapp.com>
References: <52CDAAF7.2060307@mti-systems.com> <CEF2FAA2.4E02E%prenatar@cisco.com>
In-Reply-To: <CEF2FAA2.4E02E%prenatar@cisco.com>
Accept-Language: de-AT, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.104.60.118]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: Wesley Eddy <wes@mti-systems.com>
Subject: Re: [aqm] WG status
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 10 Jan 2014 14:20:04 -0000

Preethi, Naeem, David,

thank you for stepping up!

I propose that you start off brainstorming a rough structure for the docume=
nt, such that the WG can see if all relevant aspects are addressed.

We would like to reserve a large chunk of the upcoming meeting time to disc=
uss this 2nd milestone document, after discussing the latest changes to htt=
p://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/=20

Best regards,

Richard Scheffenegger


> -----Original Message-----
> From: aqm [mailto:aqm-bounces@ietf.org] On Behalf Of Preethi Natarajan
> Sent: Mittwoch, 08. J=E4nner 2014 21:53
> To: Wesley Eddy; aqm@ietf.org
> Subject: Re: [aqm] WG status
>=20
>=20
>=20
> On 1/8/14 11:45 AM, "Wesley Eddy" <wes@mti-systems.com> wrote:
>=20
> >
> >  - Submit AQM algorithm evaluation guidelines to IESG for publication
> >as Informational (Goal: July 2014)
> >    - We need an editor team to step forward and begin work on this;
> >there was initial work presented in Vancouver, but no draft available
> >or adopted by the working group yet.
> >    - It will be difficult to make this milestone, and will push other
> >milestones back, if this work isn't accelerated.
> >    - Please express interest to the chairs or on-list
>=20
>=20
> Interested in contributing towards the evaluation guidelines draft --
> write, edit, whatever else is needed to meet deadline.
>=20
> Thanks,
> Preethi (on behalf of PIE team)
>=20
>=20
>=20
> _______________________________________________
> aqm mailing list
> aqm@ietf.org
> https://www.ietf.org/mailman/listinfo/aqm

From lars@netapp.com  Fri Jan 17 00:46:51 2014
Return-Path: <lars@netapp.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A2BCF1ACCFA for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 00:46:51 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.44
X-Spam-Level: 
X-Spam-Status: No, score=-2.44 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, RP_MATCHES_RCVD=-0.538, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id H9j6iO7LZL8v for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 00:46:50 -0800 (PST)
Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id 2EE3F1ACCEE for <aqm@ietf.org>; Fri, 17 Jan 2014 00:46:50 -0800 (PST)
X-IronPort-AV: E=Sophos;i="4.95,670,1384329600";  d="asc'?scan'208";a="96583622"
Received: from vmwexceht03-prd.hq.netapp.com ([10.106.76.241]) by mx11-out.netapp.com with ESMTP; 17 Jan 2014 00:46:37 -0800
Received: from SACEXCMBX06-PRD.hq.netapp.com ([169.254.9.60]) by vmwexceht03-prd.hq.netapp.com ([10.106.76.241]) with mapi id 14.03.0123.003; Fri, 17 Jan 2014 00:46:37 -0800
From: "Eggert, Lars" <lars@netapp.com>
To: "aqm@ietf.org" <aqm@ietf.org>
Thread-Topic: Nugget from IETF-1 proceedings
Thread-Index: AQHPE2Ci5Fh21doMxEKyKy0R6/6xKw==
Date: Fri, 17 Jan 2014 08:46:37 +0000
Message-ID: <BC39A23A-1A4F-4B24-8CC5-9A63ECF6B650@netapp.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
x-originating-ip: [10.104.60.115]
Content-Type: multipart/signed; boundary="Apple-Mail=_60E24364-A30B-4C4F-A27C-4103A4FEDC5B"; protocol="application/pgp-signature"; micalg=pgp-sha1
MIME-Version: 1.0
Cc: Ross Callon <rcallon@juniper.net>, Lixia Zhang <lixia@cs.ucla.edu>
Subject: [aqm] Nugget from IETF-1 proceedings
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 17 Jan 2014 08:46:51 -0000

--Apple-Mail=_60E24364-A30B-4C4F-A27C-4103A4FEDC5B
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=windows-1252

Nothing new under the sun. Proceedings from the very first IETF meeting, =
held 28 years ago today:

"Nagle presented his 'fair queuing' scheme, in which gateways maintain =
separate queues for each sending host. In this way, hosts with =
pathological implementations can not usurp more than their fair share of =
the gateway=92s resources. This invoked spirited and interested =
discussion. Zhang pointed out that this was a subtle change in =
architecture away from a pure datagram network. Callon reminded everyone =
that he had written a paper advocating a connection oriented Internet =
Protocol several years ago."

https://www.ietf.org/proceedings/01.pdf, page 5

Lars

--Apple-Mail=_60E24364-A30B-4C4F-A27C-4103A4FEDC5B
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment; filename="signature.asc"
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Message signed with OpenPGP using GPGMail

-----BEGIN PGP SIGNATURE-----

iQCVAwUBUtjt69ZcnpRveo1xAQI7uwP/U/KO5p6r6QA2/4WQGdYgoIGVYeXlz+72
H5ChFlQobC6xgWawQnxz1ZRS+q+u/skr9i2tnvqHCShDn50O22SEaUEeOGx19lRy
3gwSWoRUVwyJ8PJB7zHPf/KKUPt3HK/Skx/Fh5sOJCoouekzhjRXvWfHIyiao16G
OEFTQvEYhVs=
=482s
-----END PGP SIGNATURE-----

--Apple-Mail=_60E24364-A30B-4C4F-A27C-4103A4FEDC5B--

From nicolas.kuhn@telecom-bretagne.eu  Fri Jan 17 01:00:08 2014
Return-Path: <nicolas.kuhn@telecom-bretagne.eu>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C754A1AD935 for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 01:00:08 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.551
X-Spam-Level: 
X-Spam-Status: No, score=-1.551 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HELO_EQ_FR=0.35, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1lg-nGwnjem5 for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 01:00:06 -0800 (PST)
Received: from zproxy220.enst-bretagne.fr (zproxy220.enst-bretagne.fr [192.108.117.9]) by ietfa.amsl.com (Postfix) with ESMTP id 1B0601ADF2F for <aqm@ietf.org>; Fri, 17 Jan 2014 01:00:06 -0800 (PST)
Received: from localhost (localhost [127.0.0.1]) by zproxy220.enst-bretagne.fr (Postfix) with ESMTP id 312DEB0115 for <aqm@ietf.org>; Fri, 17 Jan 2014 09:59:53 +0100 (CET)
X-Virus-Scanned: amavisd-new at zproxy220.enst-bretagne.fr
Received: from zproxy220.enst-bretagne.fr ([127.0.0.1]) by localhost (zproxy220.enst-bretagne.fr [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Dl4EXyM9GE7A for <aqm@ietf.org>; Fri, 17 Jan 2014 09:59:52 +0100 (CET)
Received: from dhcp-salsa-i-129-163.rennes.enst-bretagne.fr (dhcp-salsa-i-129-163.rennes.enst-bretagne.fr [10.35.129.163]) by zproxy220.enst-bretagne.fr (Postfix) with ESMTPSA id BA2AAB00DA for <aqm@ietf.org>; Fri, 17 Jan 2014 09:59:52 +0100 (CET)
From: Nicolas KUHN <nicolas.kuhn@telecom-bretagne.eu>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: quoted-printable
Message-Id: <92D3FF75-D72D-4044-8796-E6B2A098DF5A@telecom-bretagne.eu>
Date: Fri, 17 Jan 2014 09:59:52 +0100
To: aqm@ietf.org
Mime-Version: 1.0 (Mac OS X Mail 6.3 \(1503\))
X-Mailer: Apple Mail (2.1503)
Subject: [aqm]  Evalution guidelines - Editor team
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 17 Jan 2014 09:00:09 -0000

Hi all,=20

I volunteer  to act as editor & contributor to the "AQM algorithm =
evaluation guidelines".  I just joined the RITE project [0].=20

As explained in the email of Wes Eddy to the list [1], there was initial =
work presented in Vancouver [2] and we will write a draft out of it. =20

>> Submit AQM algorithm evaluation guidelines to IESG for publication
>> as Informational (Goal: July 2014)
>>   - We need an editor team to step forward and begin work on this;
>> there was initial work presented in Vancouver, but no draft available =
or
>> adopted by the working group yet.


A draft is coming really soon: we expect to send it by mid-February.=20

Nicolas KUHN
IMT

[0] http://riteproject.eu
[1] http://www.ietf.org/mail-archive/web/aqm/current/msg00461.html
[2] http://www.ietf.org/proceedings/88/slides/slides-88-aqm-5.pdf=

From dave.taht@gmail.com  Fri Jan 17 05:29:06 2014
Return-Path: <dave.taht@gmail.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6434E1AE0BE for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 05:29:06 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Kkfz2cuq-5h7 for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 05:29:04 -0800 (PST)
Received: from mail-wi0-x234.google.com (mail-wi0-x234.google.com [IPv6:2a00:1450:400c:c05::234]) by ietfa.amsl.com (Postfix) with ESMTP id 59BBA1AE0B5 for <aqm@ietf.org>; Fri, 17 Jan 2014 05:29:04 -0800 (PST)
Received: by mail-wi0-f180.google.com with SMTP id d13so678932wiw.13 for <aqm@ietf.org>; Fri, 17 Jan 2014 05:28:51 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=6Og15M/npLAATE4Y6HwNMM5zSOufM0L3NxLYPrInh1M=; b=SIUShzxvU9Kru2DmMnWxAi69Zdwx739ft1WYysEeDPl3ErlStk6xdXjevNo8Ul+6VW rDjLMO8wOtXQ0ho/1PIIo0+cJWSyL/dmWqHymZ1t1TJpT3Sqk0Qb5OdcmJG6CHsgwm5s S7HYTeS88kUF36OQho/SASpqBn0NnKHtVdHa04YMrZmkae+vE3RwA8+YPKIXcmhekGbC BjKp+MAGw8JqwSnX9KFhJrMoHUr+DdRIqBdpb0rgcczrOaq9vUrwF88GH5ZnaL2PjEVR 1NjOoOeJv4KQcrmT0vd2p8hOqq3beTuK5G7R7ZLoIk0ek6kg8pUjiKJpmW5Ur1CEsMV8 kiVw==
MIME-Version: 1.0
X-Received: by 10.180.211.39 with SMTP id mz7mr2618366wic.53.1389965331376; Fri, 17 Jan 2014 05:28:51 -0800 (PST)
Received: by 10.217.123.69 with HTTP; Fri, 17 Jan 2014 05:28:51 -0800 (PST)
In-Reply-To: <BC39A23A-1A4F-4B24-8CC5-9A63ECF6B650@netapp.com>
References: <BC39A23A-1A4F-4B24-8CC5-9A63ECF6B650@netapp.com>
Date: Fri, 17 Jan 2014 08:28:51 -0500
Message-ID: <CAA93jw71vVCzhOUqzdE-omSaxMYq6wqmTZaaj0F5TPcj+qP__A@mail.gmail.com>
From: Dave Taht <dave.taht@gmail.com>
To: "Eggert, Lars" <lars@netapp.com>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable
Cc: Ross Callon <rcallon@juniper.net>, Alaric Moore <alaricmeister@gmail.com>, bloat <bloat@lists.bufferbloat.net>, "aqm@ietf.org" <aqm@ietf.org>, Lixia Zhang <lixia@cs.ucla.edu>
Subject: Re: [aqm] Nugget from IETF-1 proceedings
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 17 Jan 2014 13:29:06 -0000

On Fri, Jan 17, 2014 at 3:46 AM, Eggert, Lars <lars@netapp.com> wrote:
> Nothing new under the sun. Proceedings from the very first IETF meeting, =
held 28 years ago today:
>
> "Nagle presented his 'fair queuing' scheme, in which gateways maintain se=
parate queues for each sending host. In this way, hosts with pathological i=
mplementations can not usurp more than their fair share of the gateway=92s =
resources. This invoked spirited and interested discussion. Zhang pointed o=
ut that this was a subtle change in architecture away from a pure datagram =
network. Callon reminded everyone that he had written a paper advocating a =
connection oriented Internet Protocol several years ago."
>
> https://www.ietf.org/proceedings/01.pdf, page 5

It actually starts at page 88.

Reading the rest was great. Wow. What a wonderful piece of history!
Thank you for digging that up!

Seeing the problems they were trying to solve at the time was very revealin=
g.

"1000 nodes running at 9600 baud, with one outage every 5 minutes will
only use 20% of the bandwidth for routing information"

And nagles preso retains many cogent points, although I guess "source
quench" was all the rage at the time.



> Lars
>
> _______________________________________________
> aqm mailing list
> aqm@ietf.org
> https://www.ietf.org/mailman/listinfo/aqm
>



--=20
Dave T=E4ht

Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.=
html

From davec-b@rogers.com  Fri Jan 17 15:51:32 2014
Return-Path: <davec-b@rogers.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5CB421AD190 for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 15:51:32 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 0.701
X-Spam-Level: 
X-Spam-Status: No, score=0.701 tagged_above=-999 required=5 tests=[BAYES_50=0.8, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id H4XJLB4czBON for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 15:51:29 -0800 (PST)
Received: from nm25-vm1.access.bullet.mail.bf1.yahoo.com (nm25-vm1.access.bullet.mail.bf1.yahoo.com [216.109.115.192]) by ietfa.amsl.com (Postfix) with ESMTP id 1900B1ACCF8 for <aqm@ietf.org>; Fri, 17 Jan 2014 15:51:29 -0800 (PST)
Received: from [66.196.81.155] by nm25.access.bullet.mail.bf1.yahoo.com with NNFMP; 17 Jan 2014 23:51:16 -0000
Received: from [98.139.244.50] by tm1.access.bullet.mail.bf1.yahoo.com with NNFMP; 17 Jan 2014 23:51:16 -0000
Received: from [127.0.0.1] by smtp112.sbc.mail.bf1.yahoo.com with NNFMP; 17 Jan 2014 23:51:16 -0000
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rogers.com; s=s1024; t=1390002676; bh=pQumvMCPKrRIG1+muKQd/NKbXrd4g9e0laQjLTCWxNI=; h=X-Yahoo-Newman-Id:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:X-Rocket-Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:Subject:X-Enigmail-Version:Content-Type; b=X0TdJ6jHnhbWl8FUCIKc8u4cBGNh7vYxT19jROnxJBTPPHMHpn+J3CoDEC/dZs4HklYdDR7LpcYYcKcw3ildQ814a1vQ5jiyO+a120gRsFcd8hbJf4To/FMaPKAS04kaci7N9INFl27igUQuapV2tkfzLTcnfHbvOb19mMZOpvM=
X-Yahoo-Newman-Id: 199166.73888.bm@smtp112.sbc.mail.bf1.yahoo.com
X-Yahoo-Newman-Property: ymail-3
X-YMail-OSG: xhZhgJgVM1nd6qzbI4TP5yewssZA8xM.1ju4bRG2bkISMjK caxrMw1UFnkLvEtvQFJSZw6aFdmCuL3lnJ4fZJXVwlU3P89Ubp4o.Eed32qR ODdlFfuj1PDwXRPE0Cpcoi2MRbFNwQSpO7zcjvJSQfKf5vLuYUju3eqHyhh9 DrOg2crCU.KOg_7_S0uieFtEIkL0P4KfdjqWYT7K5MAp_.YaVVdV5_8dKojg thOfFy31cNydG36AqgCIA1b_792hps99BstmeXj8hiRX.90sf06_Bb.EKu86 P1Eg_PBaL7OfSElryQjlvCDHghdIhVtj5KgA3350B9aX3D.yPdu2abX.01Jc qBX4f3po5ZziIrZUQ4RrCe31LsDM4pGOuTySJWJ50i.WjcpMoH3HOJnNmQ9g fQMXEMglcvmST1ERV7z.2d9uSrU.pNl602wQdv_bAeINgI0cJaxc6Lw596Fw Cc.wHxFW_Siqur5yArF878VAl8iYuLe4ddzC0M9pPov5kjk.IUdqMba4Uush 0DRSNE.sCBLIO_3babLHgQHUrgasPmDl4JgvnlQuQhFn5YAaMHcbhlpaGLB_ lYvgz.au9lv8chX4aR2eoWXeAzX32gXmgd2heQZcsNZaTEtSjLHLSG8YE_TG Ti3OQWUReL8FdSSN534DwGeXeIcy1Mhz6CWw_WZ0oVQ5l.P7FRa6Edm8-
X-Yahoo-SMTP: sltvjZWswBCRD.ElTuB1l9j6s9wRYPpuyTNWOE5oEg--
X-Rocket-Received: from [192.168.0.12] (davec-b@99.225.150.135 with plain [98.139.221.125]) by smtp112.sbc.mail.bf1.yahoo.com with SMTP; 17 Jan 2014 15:51:16 -0800 PST
Message-ID: <52D9C1F3.3070803@rogers.com>
Date: Fri, 17 Jan 2014 18:51:15 -0500
From: David Collier-Brown <davec-b@rogers.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7
MIME-Version: 1.0
To: "aqm@ietf.org" <aqm@ietf.org>, Dave Taht <dave.taht@gmail.com>
X-Enigmail-Version: 1.6
Content-Type: multipart/alternative; boundary="------------040807090205020305090609"
Subject: [aqm] Prefatory comments re draft-aqm-reccommendation and -evaluation, and a question
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: davecb@spamcop.net
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 17 Jan 2014 23:51:32 -0000

This is a multi-part message in MIME format.
--------------040807090205020305090609
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

I've been reading through the internet-drafts, and one paragraph struck
me as very illuminating.  This is therefor a sanity-check before I go
full-hog down a particular path...

The comment is from Baker and Fairhurst,
https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/ and the
paragraph is [emphases added]

    The point of buffering in the network is to absorb data bursts and
    to transmit them during the (hopefully) ensuing bursts of silence.
    This is essential to permit the transmission of bursty data.
    Normally small queues are preferred in network devices, with
    sufficient queue capacity to absorb the bursts. The
    counter-intuitive result is that maintaining normally-small queues
    can result in higher throughput as well as lower end-to- end delay.
    In summary, queue limits should not reflect the steady state queues
    we want to be maintained in the network; i/nstead, they should
    reflect the size of bursts that a network device needs to absorb. /

All of a sudden we're talking about the kinds of queues I know a little
about (:-))
---

I'm going to suggest that these are queues and associated physical
buffers that do two things:

 1. hold packets that arrive at a bottleneck for a long as it takes to
    send them out a slower link that they came in on, and
 2. hold bursts of packets that arrive adjacent to each other until they
    can be sent out in a normal spacing, with some small amount of time
    between them


In an illustration of Dave Taht's, the first looks something like this

-------------------------+
    |X|            |X|      +-------------------
    |X|            |X|         |XXX|    |XXX|
    |X|            |X|      +-------------------
-------------------------+

At the choke-point there is a buffer at least big enough to give the
packet a chance to wheel from line into column (:-)) and start down the
smaller pipe.

The speed at which the acks come back,  the frequency of drops, and any
explicit congestion notifications slows the sender until they don't
overload the skinnier pipe, thus spacing the packets in the fatter pipe out.

Various causes [Leland] can slow or speed the packets in the fat pipe,
making it possible for several to arrive adjacent to each other,
followed by a gap.  The second purpose of a buffer is to hold  these
bursts while things space themselves back out.

They need to be big enough at minimum to do the speed matching,  and at
maximum,  big enough to spread a burst back into a normal progression,
always assuming that acks, drops and explicit congestion notifications
are slowing the sender to the speed of the slowest part of the network.
---

If I'm right about this, we can draw some helpful conclusions

  * buffer sizes can be set based on measurements:
      o speed differences, which are pretty static, plus
      o observed burstyness
  * drops and ECN can be done to match the slowest speed in the path

The latter suddenly sounds a bit like path MTU discovery, except it's a
bit more dynamic, and varies with both path and what's happening in
various parts of it.

To me, as a capacity/performance nerd, this sounds a lot more familiar
and manageable. My question to you, before I start madly scribbling on
the internet drafts is:

    /Am I blowing smoke?/

--dave

-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
davecb@spamcop.net           |                      -- Mark Twain
(416) 223-8968


--------------040807090205020305090609
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    I've been reading through the internet-drafts, and one paragraph
    struck me as very illuminating.&nbsp; This is therefor a sanity-check
    before I go full-hog down a particular path...<br>
    <br>
    The comment is from Baker and Fairhurst,
    <a class="moz-txt-link-freetext" href="https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/">https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/</a> and
    the paragraph is [emphases added]<br>
    <blockquote> The point of buffering in the network is to absorb data
      bursts and to transmit them during the (hopefully) ensuing bursts
      of silence. This is essential to permit the transmission of bursty
      data. Normally small queues are preferred in network devices, with
      sufficient queue capacity to absorb the bursts. The
      counter-intuitive result is that maintaining normally-small queues
      can result in higher throughput as well as lower end-to- end
      delay. In summary, queue limits should not reflect the steady
      state queues we want to be maintained in the network; i<i>nstead,
        they should reflect the size of bursts that a network device
        needs to absorb.
      </i><br>
    </blockquote>
    All of a sudden we're talking about the kinds of queues I know a
    little about (:-))<br>
    ---<br>
    <br>
    I'm going to suggest that these are queues and associated physical
    buffers that do two things:<br>
    <ol>
      <li>hold packets that arrive at a bottleneck for a long as it
        takes to send them out a slower link that they came in on, and </li>
      <li>hold bursts of packets that arrive adjacent to each other
        until they can be sent out in a normal spacing, with some small
        amount of time between them</li>
    </ol>
    <br>
    In an illustration of Dave Taht's, the first looks something like
    this<br>
    <br>
    -------------------------+<br>
    &nbsp;&nbsp;&nbsp; |X|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; |X|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; +-------------------<br>
    &nbsp;&nbsp;&nbsp; |X|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; |X|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; |XXX|&nbsp;&nbsp;&nbsp; |XXX|<br>
    &nbsp;&nbsp;&nbsp; |X|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; |X|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; +-------------------<br>
    -------------------------+<br>
    <br>
    At the choke-point there is a buffer at least big enough to give the
    packet a chance to wheel from line into column (:-)) and start down
    the smaller pipe.<br>
    <br>
    The speed at which the acks come back,&nbsp; the frequency of drops, and
    any explicit congestion notifications slows the sender until they
    don't overload the skinnier pipe, thus spacing the packets in the
    fatter pipe out.<br>
    <br>
    Various causes [Leland] can slow or speed the packets in the fat
    pipe, making it possible for several to arrive adjacent to each
    other, followed by a gap.&nbsp; The second purpose of a buffer is to
    hold&nbsp; these bursts while things space themselves back out.<br>
    <br>
    They need to be big enough at minimum to do the speed matching,&nbsp; and
    at maximum,&nbsp; big enough to spread a burst back into a normal
    progression, always assuming that acks, drops and explicit
    congestion notifications are slowing the sender to the speed of the
    slowest part of the network.<br>
    ---<br>
    <br>
    If I'm right about this, we can draw some helpful conclusions<br>
    <ul>
      <li>buffer sizes can be set based on measurements:</li>
      <ul>
        <li>speed differences, which are pretty static, plus</li>
        <li>observed burstyness<br>
        </li>
      </ul>
      <li>drops and ECN can be done to match the slowest speed in the
        path<br>
      </li>
    </ul>
    <p>The latter suddenly sounds a bit like path MTU discovery, except
      it's a bit more dynamic, and varies with both path and what's
      happening in various parts of it.<br>
    </p>
    <p>To me, as a capacity/performance nerd, this sounds a lot more
      familiar and manageable. My question to you, before I start madly
      scribbling on the internet drafts is:<br>
    </p>
    <p>&nbsp;&nbsp;&nbsp; <i>Am I blowing smoke?</i><br>
    </p>
    <p>--dave<br>
    </p>
    <pre class="moz-signature" cols="72">-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
<a class="moz-txt-link-abbreviated" href="mailto:davecb@spamcop.net">davecb@spamcop.net</a>           |                      -- Mark Twain
(416) 223-8968
</pre>
  </body>
</html>

--------------040807090205020305090609--

From touch@isi.edu  Fri Jan 17 16:44:40 2014
Return-Path: <touch@isi.edu>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EE2F21AD66B for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 16:44:40 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.738
X-Spam-Level: 
X-Spam-Status: No, score=-4.738 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.538] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 7odgQK239HXo for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 16:44:40 -0800 (PST)
Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by ietfa.amsl.com (Postfix) with ESMTP id 03C661AC4C1 for <aqm@ietf.org>; Fri, 17 Jan 2014 16:44:39 -0800 (PST)
Received: from [128.9.184.151] ([128.9.184.151]) (authenticated bits=0) by boreas.isi.edu (8.13.8/8.13.8) with ESMTP id s0I0i1K4017877 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Fri, 17 Jan 2014 16:44:04 -0800 (PST)
Message-ID: <52D9CE51.5060601@isi.edu>
Date: Fri, 17 Jan 2014 16:44:01 -0800
From: Joe Touch <touch@isi.edu>
User-Agent: Mozilla/5.0 (Windows NT 6.3; WOW64; rv:24.0) Gecko/20100101 Thunderbird/24.2.0
MIME-Version: 1.0
To: davecb@spamcop.net, "aqm@ietf.org" <aqm@ietf.org>, Dave Taht <dave.taht@gmail.com>
References: <52D9C1F3.3070803@rogers.com>
In-Reply-To: <52D9C1F3.3070803@rogers.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-ISI-4-43-8-MailScanner: Found to be clean
X-MailScanner-From: touch@isi.edu
Subject: Re: [aqm] Prefatory comments re draft-aqm-reccommendation and -evaluation, and a question
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 18 Jan 2014 00:44:41 -0000

On 1/17/2014 3:51 PM, David Collier-Brown wrote:
> I'm going to suggest that these are queues and associated physical
> buffers that do two things:
>
>  1. hold packets that arrive at a bottleneck for a long as it takes to
>     send them out a slower link that they came in on, and
>  2. hold bursts of packets that arrive adjacent to each other until they
>     can be sent out in a normal spacing, with some small amount of time
>     between them

That's one thing:

	hold packets that have already arrived until they can depart

If you're talking about rate changes, then the queue might absorb a 
burst of packets then emit them, i.e., you need a queue that can handle 
several packets.

If you're taking about scheduling, then a queue might need as little as 
one packet to adjust the outgoing time based on the incoming time.

However, they're both queues, and they both absorb "bursts" (in the 
latter case, it's a 'burst' of 1 that needs to wait until a scheduled 
slot is available).

Joe

From davec-b@rogers.com  Fri Jan 17 17:42:13 2014
Return-Path: <davec-b@rogers.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 14CAC1AC4A3 for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 17:42:13 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id JRHgaczyQNXh for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 17:42:11 -0800 (PST)
Received: from nm15-vm4.access.bullet.mail.bf1.yahoo.com (nm15-vm4.access.bullet.mail.bf1.yahoo.com [216.109.115.35]) by ietfa.amsl.com (Postfix) with ESMTP id 3FF321A1F74 for <aqm@ietf.org>; Fri, 17 Jan 2014 17:42:11 -0800 (PST)
Received: from [66.196.81.156] by nm15.access.bullet.mail.bf1.yahoo.com with NNFMP; 18 Jan 2014 01:41:58 -0000
Received: from [98.139.244.51] by tm2.access.bullet.mail.bf1.yahoo.com with NNFMP; 18 Jan 2014 01:41:58 -0000
Received: from [127.0.0.1] by smtp113.sbc.mail.bf1.yahoo.com with NNFMP; 18 Jan 2014 01:41:58 -0000
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rogers.com; s=s1024; t=1390009318; bh=W+GjU5AViE2t3jsn/VfCMEKhDX5C9eK1Go3KUiqW4lE=; h=X-Yahoo-Newman-Id:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:X-Rocket-Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:Subject:References:In-Reply-To:X-Enigmail-Version:Content-Type:Content-Transfer-Encoding; b=xzUb0Ka/lfNltc8S8yozODyvUBPG9mU26l5sWyq1MMIELGTRdHSY7GWXgbTRODDC4AFUNO2zHbJJTBl7/26KBjHL2M/ahyppoxa6B2b4oJrtBzcTHTPTa6u8uEVmEwnZbofzNxsJFhTomSOmF7tOphCXIg2QAxfErUmSshSseSM=
X-Yahoo-Newman-Id: 544556.32721.bm@smtp113.sbc.mail.bf1.yahoo.com
X-Yahoo-Newman-Property: ymail-3
X-YMail-OSG: P6ffRWsVM1kYR4W3.LMsHC2zX_s8yD5aKZWnIAKxUgj8vkr 8.v77J5rMIEumYQy_MdGxFEoRvafDsbJ9UpD55SwiwVP_bZU_y.P9gf1H7I2 bLeJOW.7QAd1tS4E7Oj3QnP767sQojM4otfV4FAKVivW52jvQy1D1RJ0bl0U nabcdCZRmlWYmnYknv1iPpelmSxCw6QxHzSLAVX5BnRoAhsiZLiqJ1EUblMx ZtxEoQaxKIFySJDBVqDR5HK2MpvlAyaBRmFCmHArmuPCnF2JiDfdgxuFkzwx RHwRMberPIn.0HVEGqDXHZZVeMf3MGs2fE1qf9KyA2lXGyJ86mmwReAmpdpc gEnwbG6kraZ4AlbtBVQve3OpwdJTw1H87disU6SwVncX_YK.2gcoG6vO86Gg t8GfnF_pBHI3hBoXNVGp9HTbXmE_vMixmnbzgzrn2HNLSnXHrvVTE8raTxV9 ZLqRHHbK2YS0B8yl2ceNALT47Se4K9LD.A0RjSwMm3yzloZgznqLvPeYgXr4 M9UqM_GcesGf6UAe95vsjZW.UvURRA8Qq.rhUmtlaDMj9PxegjFRpMICKr1h 3Pn4-
X-Yahoo-SMTP: sltvjZWswBCRD.ElTuB1l9j6s9wRYPpuyTNWOE5oEg--
X-Rocket-Received: from [192.168.0.12] (davec-b@99.225.150.135 with plain [98.139.221.125]) by smtp113.sbc.mail.bf1.yahoo.com with SMTP; 17 Jan 2014 17:41:57 -0800 PST
Message-ID: <52D9DBE4.7020601@rogers.com>
Date: Fri, 17 Jan 2014 20:41:56 -0500
From: David Collier-Brown <davec-b@rogers.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7
MIME-Version: 1.0
To: aqm@ietf.org, Joe Touch <touch@isi.edu>
References: <52D9C1F3.3070803@rogers.com> <52D9CE51.5060601@isi.edu>
In-Reply-To: <52D9CE51.5060601@isi.edu>
X-Enigmail-Version: 1.6
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Subject: Re: [aqm] Prefatory comments re draft-aqm-reccommendation and -evaluation, and a question
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: davecb@spamcop.net
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 18 Jan 2014 01:42:13 -0000

Excellent, thanks: I had forgotten the &@%^#$##!$!!!!! pass-through
switch that we had a few years ago, that didn't have enough buffer and
threw a jam signal whenever it got busy.

--dave

On 01/17/2014 07:44 PM, Joe Touch wrote:
>
>
> On 1/17/2014 3:51 PM, David Collier-Brown wrote:
>> I'm going to suggest that these are queues and associated physical
>> buffers that do two things:
>>
>>  1. hold packets that arrive at a bottleneck for a long as it takes to
>>     send them out a slower link that they came in on, and
>>  2. hold bursts of packets that arrive adjacent to each other until they
>>     can be sent out in a normal spacing, with some small amount of time
>>     between them
>
> That's one thing:
>
>     hold packets that have already arrived until they can depart
>
> If you're talking about rate changes, then the queue might absorb a
> burst of packets then emit them, i.e., you need a queue that can
> handle several packets.
>
> If you're taking about scheduling, then a queue might need as little
> as one packet to adjust the outgoing time based on the incoming time.
>
> However, they're both queues, and they both absorb "bursts" (in the
> latter case, it's a 'burst' of 1 that needs to wait until a scheduled
> slot is available).
>
> Joe
> _______________________________________________
> aqm mailing list
> aqm@ietf.org
> https://www.ietf.org/mailman/listinfo/aqm
>


-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
davecb@spamcop.net           |                      -- Mark Twain
(416) 223-8968


From davec-b@rogers.com  Sat Jan 18 06:06:19 2014
Return-Path: <davec-b@rogers.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 187001ADE7C for <aqm@ietfa.amsl.com>; Sat, 18 Jan 2014 06:06:19 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id yKkQ_hgllNVQ for <aqm@ietfa.amsl.com>; Sat, 18 Jan 2014 06:06:17 -0800 (PST)
Received: from nm26-vm5.access.bullet.mail.bf1.yahoo.com (nm26-vm5.access.bullet.mail.bf1.yahoo.com [216.109.115.212]) by ietfa.amsl.com (Postfix) with ESMTP id 98B941ADDD2 for <aqm@ietf.org>; Sat, 18 Jan 2014 06:06:17 -0800 (PST)
Received: from [66.196.81.165] by nm26.access.bullet.mail.bf1.yahoo.com with NNFMP; 18 Jan 2014 14:06:04 -0000
Received: from [98.139.221.157] by tm11.access.bullet.mail.bf1.yahoo.com with NNFMP; 18 Jan 2014 14:06:04 -0000
Received: from [127.0.0.1] by smtp117.sbc.mail.bf1.yahoo.com with NNFMP; 18 Jan 2014 14:06:04 -0000
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rogers.com; s=s1024; t=1390053964; bh=qXWcUs+Rn00lNzxyCGqakjOKg5YUZScu2XF1is1CX58=; h=X-Yahoo-Newman-Id:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:X-Rocket-Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:CC:Subject:References:In-Reply-To:X-Enigmail-Version:Content-Type; b=pkin1FZLN4RV6Zy0RMzIeWXjcXBOvwyBvb7scdCV0sYkDXt6f4jjl5ZiD4v3hz8HsyVX+80rSXN4TGJJebYH3iwCn4Guzx2W5lZgQDbb65fkarhGXWiKwqivBu9u0aiTjC7Rp0WTDjz0SSW0Pt2+foo04WYQR+AQqzeKuzPX638=
X-Yahoo-Newman-Id: 613989.8980.bm@smtp117.sbc.mail.bf1.yahoo.com
X-Yahoo-Newman-Property: ymail-3
X-YMail-OSG: BJyYdsUVM1kp.Q8e9RBmxbPvkubaBmwRNEs5wojsJh808iE LAPW7UMKlRPYcgMvcVlocghxWiqSEl5mSmx03IDaz90EmIa_f._ohZlNSGKG NjlctMcL09wNvcJkDBcQ3Q_rNtJQ9kEbcDVEf7a.qTi4mjJI_kt3EUsp.r8M U4CXVB2x4tdEsmvvJ6y02upYrkYewhYQLs4uyG_9iAweHKfDgcCl9Ab24GPe hFxbro3ldonLMmfJP43giDeIKvzVn7JmYMLgNpxA1MGnhyYxN4BFr9MhG5El tx30iHLllocJceL6MulapTALK7Nhg8c2efwDu..IuKJtisSf3IC2pnfPC_.u mM4SMW7rxNa4oBF8KdDlMOaFGGQAIIsjrp_93n7pw0ZzEWavS85aTQWDyYAa 1gqWzAL9TiyX0zXr9RltvlRd327xof8oIE.WWnH0thVJl0oHgzQLW3mPBiY3 qUkedO8jHLjeCR.kgq1rJkF5pnld3oLVnNiJh8ihiR4.lVKegC7MFhx_cKAl m4P2LZ5IwhfUWYY8d5D5JtvtDCAEAUWxbBsjHV2QGWmVaDraXWRYzsLDRWXU Knuw2YPY-
X-Yahoo-SMTP: sltvjZWswBCRD.ElTuB1l9j6s9wRYPpuyTNWOE5oEg--
X-Rocket-Received: from [192.168.0.12] (davec-b@99.225.150.135 with plain [98.139.221.125]) by smtp117.sbc.mail.bf1.yahoo.com with SMTP; 18 Jan 2014 14:06:04 +0000 UTC
Message-ID: <52DA9A49.9040109@rogers.com>
Date: Sat, 18 Jan 2014 10:14:17 -0500
From: David Collier-Brown <davec-b@rogers.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7
MIME-Version: 1.0
To: davecb@spamcop.net
References: <52D9C1F3.3070803@rogers.com>
In-Reply-To: <52D9C1F3.3070803@rogers.com>
X-Enigmail-Version: 1.6
Content-Type: multipart/alternative; boundary="------------050807040707060805070206"
Cc: "aqm@ietf.org" <aqm@ietf.org>
Subject: Re: [aqm] Prefatory comments re draft-aqm-reccommendation and -evaluation, and a question
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: davecb@spamcop.net
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 18 Jan 2014 14:06:19 -0000

This is a multi-part message in MIME format.
--------------050807040707060805070206
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

On 01/17/2014 06:51 PM, I wrote:
> I've been reading through the internet-drafts, and one paragraph
> struck me as very illuminating.  This is therefor a sanity-check
> before I go full-hog down a particular path...

I have a longish set of comments for draft-ietf-aqm-recommendations-00,
is it best they go to the list as a series, or to a smaller group as one
big wodge?

--dave

-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
davecb@spamcop.net           |                      -- Mark Twain
(416) 223-8968


--------------050807040707060805070206
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">On 01/17/2014 06:51 PM, I wrote:<br>
    </div>
    <blockquote cite="mid:52D9C1F3.3070803@rogers.com" type="cite">
      <meta http-equiv="content-type" content="text/html;
        charset=ISO-8859-1">
      I've been reading through the internet-drafts, and one paragraph
      struck me as very illuminating.&nbsp; This is therefor a sanity-check
      before I go full-hog down a particular path...<br>
    </blockquote>
    <br>
    I have a longish set of comments for
    draft-ietf-aqm-recommendations-00, is it best they go to the list as
    a series, or to a smaller group as one big wodge?<br>
    <br>
    --dave<br>
    <pre class="moz-signature" cols="72">-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
<a class="moz-txt-link-abbreviated" href="mailto:davecb@spamcop.net">davecb@spamcop.net</a>           |                      -- Mark Twain
(416) 223-8968
</pre>
  </body>
</html>

--------------050807040707060805070206--

From davec-b@rogers.com  Sat Jan 18 17:45:21 2014
Return-Path: <davec-b@rogers.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 86D281AD67A for <aqm@ietfa.amsl.com>; Sat, 18 Jan 2014 17:45:21 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 0.701
X-Spam-Level: 
X-Spam-Status: No, score=0.701 tagged_above=-999 required=5 tests=[BAYES_50=0.8, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id C0CVjDkEFaBc for <aqm@ietfa.amsl.com>; Sat, 18 Jan 2014 17:45:18 -0800 (PST)
Received: from nm18-vm6.access.bullet.mail.bf1.yahoo.com (nm18-vm6.access.bullet.mail.bf1.yahoo.com [216.109.115.85]) by ietfa.amsl.com (Postfix) with ESMTP id 492CE1AD190 for <aqm@ietf.org>; Sat, 18 Jan 2014 17:45:18 -0800 (PST)
Received: from [66.196.81.166] by nm18.access.bullet.mail.bf1.yahoo.com with NNFMP; 19 Jan 2014 01:45:05 -0000
Received: from [98.139.244.50] by tm12.access.bullet.mail.bf1.yahoo.com with NNFMP; 19 Jan 2014 01:45:05 -0000
Received: from [127.0.0.1] by smtp112.sbc.mail.bf1.yahoo.com with NNFMP; 19 Jan 2014 01:45:05 -0000
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rogers.com; s=s1024; t=1390095905; bh=SLHf7qNi/4guBkjUSv19ubbA1dHDOkioGZ3T57Nawds=; h=X-Yahoo-Newman-Id:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:X-Rocket-Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:Subject:X-Enigmail-Version:Content-Type; b=dljKf4c4GwkTc3Wl5+WNJL7ynunC7pCe3sPD6R4rIUKY8sbg9huQp7XVMaPOXybr90UKb8G+Ek3vNx0uhNeGRvf/hAhkGKVvBBERaBLaSFnm0g2nlpNlMN4BzTjO14g47omUnHTcfpktzi2OWiddJmVCJ6lWp6vNtE4ShJyx5Ns=
X-Yahoo-Newman-Id: 94513.62475.bm@smtp112.sbc.mail.bf1.yahoo.com
X-Yahoo-Newman-Property: ymail-3
X-YMail-OSG: 5rihlYUVM1nXsDaO_ZN4UMNOuW6PYJn0pcZNqF5oKBc5k24 b4K_jZ6isFTFziP.Xb5oZYKTpH0DTemaXM1gnjQOHDpPcMFLKE5SsCzDosO5 eC21TzGKVedqBQn.u0jpWmAQKPCxlAxeJ6HXkmkTuA29fPFuqMQaOZ.AHPpF 5VrhtcmnPsS5AO.YPZ7lo0henCm19XcA9SXO373YRw9X3kZOFJVhKnoCrSZ3 q.TepVJhaJAOSqNrDTAcX8VSW8lM8.oll3s15yANDgSoNv7WfifIlqaHy2tk j_ymPOjd.pgLj_L_tIvhrWPSiBTVDaeMa0755BSK_W_gRMVzG.3rOOAOysmk 8HzaLkCVBwW6q6olcySgx8RxQ6aYvbuhfK0x3fRP7K9xRxvAkR6L6wPbVndV Qzi7hhrO6BFrq9JXj7y9y_A1rDgXx09nf6_57Px1EVZ3lA9ls4BryvWRCugx CV50yy_HN2thzmyLWJ5MyGxB7vYKe2Gr7enBAFybWrb0L.cUt3r0hELchZII G26you6aqrV1DFbr3JVlGF4xCLc44OtTOqtffOc7zBETfceq6PerdSrQ9Wv9 DO7Gql1I-
X-Yahoo-SMTP: sltvjZWswBCRD.ElTuB1l9j6s9wRYPpuyTNWOE5oEg--
X-Rocket-Received: from [192.168.0.12] (davec-b@99.225.150.135 with plain [98.139.221.125]) by smtp112.sbc.mail.bf1.yahoo.com with SMTP; 18 Jan 2014 17:45:05 -0800 PST
Message-ID: <52DB2E1F.50009@rogers.com>
Date: Sat, 18 Jan 2014 20:45:03 -0500
From: David Collier-Brown <davec-b@rogers.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7
MIME-Version: 1.0
To: "aqm@ietf.org" <aqm@ietf.org>
X-Enigmail-Version: 1.6
Content-Type: multipart/alternative; boundary="------------040204040008010603030302"
Subject: [aqm] Comments on draft-ietf-aqm-recommendation-00, message 1 of several
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: davecb@spamcop.net
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 19 Jan 2014 01:45:21 -0000

This is a multi-part message in MIME format.
--------------040204040008010603030302
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

Section, paragraph (page)
=======  =========  ====

1, 3 (p3)

   ... mechanisms [RFC5681], while necessary and powerful, are not
   sufficient to provide good service in all circumstances.  Basically,
   there is a limit to how much control can be accomplished from the
   edges of the network.  Some mechanisms are needed in the network
   devices to complement the endpoint congestion avoidance mechanisms.

This reads a bit like middleboxes aren't using anything, and that they
need different
mechanisms.  I'd suggest they need to be as active as using the same
control
mechanisms as the endpoints, or at least their components of the control
mechanisms.

How about "Network devices need to actively use the same mechanisms to
provide hop-by-hop congestion avoidance as are used by the endpoints,
especially the newer mechanisms."  That reflects the fact that the
experimentation is going on at or quite near the edge, and the middle
has been a bit left out.

I'd also suggest the second sentence would be stronger if it doesn't
begin with "basically".


1, 4 (p3)
   ... To a rough approximation, queue management algorithms
   manage the length of packet queues by marking or dropping packets
   when necessary or appropriate, while scheduling algorithms determine ...

Based on the description in 1.1.2, para 2 (the one about queues I
mentioned in the prefatory remarks), I suggest we say here that queue
management algorithms "/minimize/ the length of packet queues". 

This is a bit of a truism in the queuing world, and a different mind-set
than buffer management. One wouldn't say "minimize buffering", as that
would draw an entirely backwards picture in the reader's mind (;-)), but
we do wish to minimize the waiting-in-queue, for both throughput and
latency reasons, not just for one or the other.


1, 6 (p3)
   The second issue, discussed in Section 3 of this memo, is the
   potential for future congestive collapse of the Internet due to
   ...
   available.  It is imperative that this work be energetically pursued,
   to ensure the future stability of the Internet.

Unless it is impossible for congestive collapse to happen now, I'd
suggest putting this into the present tense: say "potential for
congestive collapse" and "retain the stability of the internet"
As it stands, it sounds like we're arguing that this can only happen at
some future date, when something worsens, and so the reader could
reasonably conclude that they need not act.

1, 8 (p4) 
   The discussion in this memo applies to "best-effort" traffic, which
   is to say, traffic generated by applications that accept the
   occasional loss, duplication, or reordering of traffic in flight.  It
   also applies to other traffic, such as real-time traffic that can
   adapt its sending rate to reduce loss and/or delay.  It is most
   effective, when the adaption occurs on time scales of a single RTT or
   a small number of RTTs, for elastic traffic [RFC1633].

I find this just plain hard to read.  I'm going to assume you mean to
identify the traffic regimes we consider applicable, and in indicate
their desirability. If so, may I suggest:

   The discussion in this memo applies to most observed kinds of traffic, but not to <specific 
   technical term goes here> ones, where no delay, loss or reordering can be permitted.  It works 
   for so-called "real-time" traffic that can adapt its sending rate to reduce loss and/or delay, 
   and is optimal for the common case, "best effort" traffic, traffic that can accept the
   occasional loss, duplication, or reordering of traffic in flight. It is most effective for 
   <a specific technical term goes here too?> traffic where such events occur on time scales of a 
   single RTT or a small number of RTTs.


I initially through you were referring forward to the responsive,
unresponsive and tcp-friendly flows you mention in section 3, but they
clearly don't fit: can you flesh this paragraph out with the kinds of
traffic you mean, or indicate if you meant something else entirely?

2, 1 (p4)

   The traditional technique for managing the queue length in a network
   device is to set a maximum length (in terms of packets) for each
   queue, accept packets for the queue until the maximum length is
   reached, then reject (drop) subsequent incoming packets until the
   queue decreases because a packet from the queue has been transmitted.

I'd break the paragraph here and let its name be a separate paragraph.

I'd follow it with a statement of why we have to deal with queues, such as

    These queues build up wherever there is a connection between an incoming and an 
    outgoing stream of packets, and a buffer is used to hold packets where they are 
    being routed, where the input and output are running at different speeds, or 
    where a group of packets arrive together at the input and they need to be held until
    output processing can drain them away.

This then distinguishes between the three uses of the buffer, and
motivates the following argument.

2.1, 1 (p4)

       In some situations tail drop allows a single connection or a few
       flows to monopolize queue space, preventing other connections
       from getting room in the queue.  This "lock-out" phenomenon is
       often the result of synchronization or other timing effects.


I think it might be better to say "inadvertent synchronization", as that
avoids suggesting that some separate and intentional synchronization was
causing the problem.

2.1, 2 (p4)

       The tail drop discipline allows queues to maintain a full (or,
       almost full) buffer for long periods of time, since tail drop
       signals congestion (via a packet drop) only when the queue has
       become full.  

I changed "status" to "buffer" in the above. Following this sentence, may I
suggest we skip immediately to

       The point of buffering in the network is to absorb data bursts
       and to transmit them during the (hopefully) ensuing bursts of
       silence.  This is essential to permit the transmission of bursty
       data.  Normally small queues are preferred in network devices,
       with sufficient queue capacity to absorb the bursts. 

then

       Even though TCP constrains the congestion window of
       a flow, packets often arrive at network devices in bursts
       [Leland94].  If the queue is full or almost full, an arriving
       burst will cause multiple packets to be dropped.  This can result
       in a global synchronization of flows throttling back, followed by
       a sustained period of lowered link utilization, reducing overall
       throughput.

       The counter-intuitive result is that maintaining normally-small
       buffer residencies (queues) can result in higher throughput as 
       well as lower end-to-end delay.  In summary, buffer/queue limits 
       should not reflect the steady state queues we want to be maintained 
       in the network; instead, they should reflect the size of bursts that 
       a network device needs to absorb. 

I Inserted "buffer residencies (queues)" in place of "queues" and
"buffer/queue limits" in place of "queue limits", then we could follow
up with

       The naive assumption might be that there is a simple tradeoff
       between delay and throughput, and that the recommendation that
       queues be maintained in a "non-full" state essentially translates
       to a recommendation that low end-to-end delay is more important
       than high throughput. 

I'd then conclude with

        As we see above, this is by no means the case. Small queues and 
        normally-empty buffers maximize throughput as well.


2.2, 5 (p5)

   We know in general how to solve the full-queues problem for
   "responsive" flows, i.e., those flows that throttle back in response
   to congestion notification.  In the current Internet, dropped packets
   provide a critical mechanism indicating congestion notification to
   hosts.  The solution to the full-queues problem is for network
   devices to drop packets before a queue becomes full, so that hosts
   can respond to congestion before buffers overflow.  We call such a
   proactive approach AQM.  By dropping packets before buffers overflow,
   AQM allows network devices to control when and how many packets to
   drop.

As we're going to use ECN fairly soon as part of AQM, we need to
foreshadow that, and this is also a good place to define AQM.  I suggest
changing  "The solution to the full-queues problem ... We call such a
proactive approach AQM. " to

   The solution to the full-queues problem is for network devices to drop packets
   or otherwise notify the sender before a queue becomes full, so that hosts
   can respond to congestion before buffers overflow.  We define such a proactive
   approach to be /Advanced Queue Management/ (AQM).  By dropping or marking packets
   before buffers overflow, AQM allows network devices to maximize performance.

End of first day: comments, suggestions and brickbats cordially accepted!

--dave
[If anyone would prefer this as amendments to an editable format, just say]

--------------040204040008010603030302
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <pre>Section, paragraph (page)
=======&nbsp; =========  ====</pre>
    <pre>1, 3 (p3)
</pre>
    <pre>   ... mechanisms [RFC5681], while necessary and powerful, are not
   sufficient to provide good service in all circumstances.  Basically,
   there is a limit to how much control can be accomplished from the
   edges of the network.  Some mechanisms are needed in the network
   devices to complement the endpoint congestion avoidance mechanisms.</pre>
    This reads a bit like middleboxes aren't using anything, and that
    they need different <br>
    mechanisms.&nbsp; I'd suggest they need to be as active as using the same
    control <br>
    mechanisms as the endpoints, or at least their components of the
    control mechanisms.<br>
    <br>
    How about "Network devices need to actively use the same mechanisms
    to provide hop-by-hop congestion avoidance as are used by the
    endpoints, especially the newer mechanisms."&nbsp; That reflects the fact
    that the experimentation is going on at or quite near the edge, and
    the middle has been a bit left out.<br>
    <br>
    I'd also suggest the second sentence would be stronger if it doesn't
    begin with "basically".<br>
    <br>
    <br>
    <pre>1, 4 (p3)
   ... To a rough approximation, queue management algorithms
   manage the length of packet queues by marking or dropping packets
   when necessary or appropriate, while scheduling algorithms determine ...

</pre>
    Based on the description in 1.1.2, para 2 (the one about queues I
    mentioned in the prefatory remarks), I suggest we say here that
    queue management algorithms "<i>minimize</i> the length of packet
    queues".&nbsp; <br>
    <br>
    This is a bit of a truism in the queuing world, and a different
    mind-set than buffer management. One wouldn't say "minimize
    buffering", as that would draw an entirely backwards picture in the
    reader's mind (;-)), but we do wish to minimize the
    waiting-in-queue, for both throughput and latency reasons, not just
    for one or the other.<br>
    <br>
    <br>
    <pre>1, 6 (p3)
   The second issue, discussed in Section 3 of this memo, is the
   potential for future congestive collapse of the Internet due to
   ...
   available.  It is imperative that this work be energetically pursued,
   to ensure the future stability of the Internet.

</pre>
    Unless it is impossible for congestive collapse to happen now, I'd
    suggest putting this into the present tense: say "potential for
    congestive collapse" and "retain the stability of the internet"<br>
    As it stands, it sounds like we're arguing that this can only happen
    at some future date, when something worsens, and so the reader could
    reasonably conclude that they need not act.<br>
    <br>
    <pre>1, 8 (p4) 
   The discussion in this memo applies to "best-effort" traffic, which
   is to say, traffic generated by applications that accept the
   occasional loss, duplication, or reordering of traffic in flight.  It
   also applies to other traffic, such as real-time traffic that can
   adapt its sending rate to reduce loss and/or delay.  It is most
   effective, when the adaption occurs on time scales of a single RTT or
   a small number of RTTs, for elastic traffic [RFC1633].</pre>
    I find this just plain hard to read.&nbsp; I'm going to assume you mean
    to identify the traffic regimes we consider applicable, and in
    indicate their desirability. If so, may I suggest:<br>
    <pre>   The discussion in this memo applies to most observed kinds of traffic, but not to &lt;specific 
   technical term goes here&gt; ones, where no delay, loss or reordering can be permitted.&nbsp; It works 
   for so-called "real-time" traffic that can adapt its sending rate to reduce loss and/or delay, 
   and is optimal for the common case, "best effort" traffic, traffic that can accept the
   occasional loss, duplication, or reordering of traffic in flight. It is most effective for 
   &lt;a specific technical term goes here too?&gt; traffic where such events occur on time scales of a 
   single RTT or a small number of RTTs.</pre>
    <br>
    I initially through you were referring forward to the responsive,
    unresponsive and tcp-friendly flows you mention in section 3, but
    they clearly don't fit: can you flesh this paragraph out with the
    kinds of traffic you mean, or indicate if you meant something else
    entirely?<br>
    <br>
    <pre>2, 1 (p4)</pre>
    <pre>   The traditional technique for managing the queue length in a network
   device is to set a maximum length (in terms of packets) for each
   queue, accept packets for the queue until the maximum length is
   reached, then reject (drop) subsequent incoming packets until the
   queue decreases because a packet from the queue has been transmitted.

</pre>
    <pre>I'd break the paragraph here and let its name be a separate paragraph.
</pre>
    I'd follow it with a statement of why we have to deal with queues,
    such as <br>
    <pre>&nbsp;&nbsp;&nbsp; These queues build up wherever there is a connection between an incoming and an 
    outgoing stream of packets, and a buffer is used to hold packets where they are 
    being routed, where the input and output are running at different speeds, or 
    where a group of packets arrive together at the input and they need to be held until
    output processing can drain them away.</pre>
    This then distinguishes between the three uses of the buffer, and
    motivates the following argument.<br>
    <br>
    2.1, 1 (p4)<br>
    <pre>       In some situations tail drop allows a single connection or a few
       flows to monopolize queue space, preventing other connections
       from getting room in the queue.  This "lock-out" phenomenon is
       often the result of synchronization or other timing effects.</pre>
    <br>
    I think it might be better to say "inadvertent synchronization", as
    that avoids suggesting that some separate and intentional
    synchronization was causing the problem.<br>
    <br>
    <pre>2.1, 2 (p4)</pre>
    <pre>       The tail drop discipline allows queues to maintain a full (or,
       almost full) buffer for long periods of time, since tail drop
       signals congestion (via a packet drop) only when the queue has
       become full.  

</pre>
    I changed "status" to "buffer" in the above. Following this
    sentence, may I <br>
    suggest we skip immediately to <br>
    <pre>       The point of buffering in the network is to absorb data bursts
       and to transmit them during the (hopefully) ensuing bursts of
       silence.  This is essential to permit the transmission of bursty
       data.  Normally small queues are preferred in network devices,
       with sufficient queue capacity to absorb the bursts. 

then

       Even though TCP constrains the congestion window of
       a flow, packets often arrive at network devices in bursts
       [Leland94].  If the queue is full or almost full, an arriving
       burst will cause multiple packets to be dropped.  This can result
       in a global synchronization of flows throttling back, followed by
       a sustained period of lowered link utilization, reducing overall
       throughput.

</pre>
    <pre>       The counter-intuitive result is that maintaining normally-small
       buffer residencies (queues) can result in higher throughput as 
       well as lower end-to-end delay.  In summary, buffer/queue limits 
       should not reflect the steady state queues we want to be maintained 
       in the network; instead, they should reflect the size of bursts that 
       a network device needs to absorb. </pre>
    I Inserted "buffer residencies (queues)" in place of "queues" and
    "buffer/queue limits" in place of "queue limits", then we could
    follow up with <br>
    <br>
    <pre>       The naive assumption might be that there is a simple tradeoff
       between delay and throughput, and that the recommendation that
       queues be maintained in a "non-full" state essentially translates
       to a recommendation that low end-to-end delay is more important
       than high throughput. 

</pre>
    I'd then conclude with <br>
    <pre>        As we see above, this is by no means the case. Small queues and 
        normally-empty buffers maximize throughput as well.</pre>
    <br>
    2.2, 5 (p5)<br>
    <pre>   We know in general how to solve the full-queues problem for
   "responsive" flows, i.e., those flows that throttle back in response
   to congestion notification.  In the current Internet, dropped packets
   provide a critical mechanism indicating congestion notification to
   hosts.  The solution to the full-queues problem is for network
   devices to drop packets before a queue becomes full, so that hosts
   can respond to congestion before buffers overflow.  We call such a
   proactive approach AQM.  By dropping packets before buffers overflow,
   AQM allows network devices to control when and how many packets to
   drop.

</pre>
    As we're going to use ECN fairly soon as part of AQM, we need to
    foreshadow that, and this is also a good place to define AQM.&nbsp; I
    suggest changing&nbsp; "The solution to the full-queues problem ... We
    call such a proactive approach AQM. " to <br>
    <pre>   The solution to the full-queues problem is for network devices to drop packets
   or otherwise notify the sender before a queue becomes full, so that hosts
   can respond to congestion before buffers overflow.  We define such a proactive
   approach to be <i>Advanced Queue Management</i> (AQM).  By dropping or marking packets
   before buffers overflow, AQM allows network devices to maximize performance.</pre>
    End of first day: comments, suggestions and brickbats cordially
    accepted!<br>
    <br>
    --dave<br>
    [If anyone would prefer this as amendments to an editable format,
    just say]<br>
  </body>
</html>

--------------040204040008010603030302--

From lixia@cs.ucla.edu  Fri Jan 17 20:57:12 2014
Return-Path: <lixia@cs.ucla.edu>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6A23C1ADBD2 for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 20:57:12 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.739
X-Spam-Level: 
X-Spam-Status: No, score=-4.739 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.538, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OJssPcbG_bFh for <aqm@ietfa.amsl.com>; Fri, 17 Jan 2014 20:57:11 -0800 (PST)
Received: from smtp.cs.ucla.edu (smtp.cs.ucla.edu [131.179.128.62]) by ietfa.amsl.com (Postfix) with ESMTP id 3DC9F1AD958 for <aqm@ietf.org>; Fri, 17 Jan 2014 20:57:11 -0800 (PST)
Received: from localhost (localhost.localdomain [127.0.0.1]) by smtp.cs.ucla.edu (Postfix) with ESMTP id AF0F2A60005; Fri, 17 Jan 2014 20:56:58 -0800 (PST)
X-Virus-Scanned: amavisd-new at smtp.cs.ucla.edu
Received: from smtp.cs.ucla.edu ([127.0.0.1]) by localhost (smtp.cs.ucla.edu [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Wi8Uf5QMochS; Fri, 17 Jan 2014 20:56:57 -0800 (PST)
Received: from [10.0.1.8] (cpe-76-166-241-160.socal.res.rr.com [76.166.241.160]) by smtp.cs.ucla.edu (Postfix) with ESMTPSA id 4E54BA60004; Fri, 17 Jan 2014 20:56:57 -0800 (PST)
Content-Type: text/plain; charset=windows-1252
Mime-Version: 1.0 (Mac OS X Mail 6.6 \(1510\))
From: Lixia Zhang <lixia@cs.ucla.edu>
In-Reply-To: <BC39A23A-1A4F-4B24-8CC5-9A63ECF6B650@netapp.com>
Date: Fri, 17 Jan 2014 20:56:57 -0800
Content-Transfer-Encoding: quoted-printable
Message-Id: <FD058BE4-4E4B-4EA7-87F1-C611B6B8A143@cs.ucla.edu>
References: <BC39A23A-1A4F-4B24-8CC5-9A63ECF6B650@netapp.com>
To: "Eggert, Lars" <lars@netapp.com>
X-Mailer: Apple Mail (2.1510)
X-Mailman-Approved-At: Sat, 18 Jan 2014 19:39:46 -0800
Cc: Ross Callon <rcallon@juniper.net>, "aqm@ietf.org" <aqm@ietf.org>
Subject: Re: [aqm] Nugget from IETF-1 proceedings
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 18 Jan 2014 04:57:12 -0000

On Jan 17, 2014, at 12:46 AM, "Eggert, Lars" <lars@netapp.com> wrote:

> Nothing new under the sun. Proceedings from the very first IETF =
meeting, held 28 years ago today:
>=20
> "Nagle presented his 'fair queuing' scheme, in which gateways maintain =
separate queues for each sending host. In this way, hosts with =
pathological implementations can not usurp more than their fair share of =
the gateway=92s resources. This invoked spirited and interested =
discussion. Zhang pointed out that this was a subtle change in =
architecture away from a pure datagram network. Callon reminded everyone =
that he had written a paper advocating a connection oriented Internet =
Protocol several years ago."
>=20
> https://www.ietf.org/proceedings/01.pdf, page 5
>=20
> Lars


thanks for digging this out!  My memory for that meeting totally faded =
away, other than the weather was gorgeous (compared to Boson;)

I got into networking in 1981, I recall a number of other Internet =
meetings before this one but I guess they left no traces...  it is great =
that people spent serious effort to keep IETF history.

Lixia
PS: about that "subtle change away from pure datagram" comment: it's =
amusing how little we/I understood the issue of keeping state inside =
network at the time.=

From davec-b@rogers.com  Sun Jan 19 14:22:13 2014
Return-Path: <davec-b@rogers.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C09751ADFA4 for <aqm@ietfa.amsl.com>; Sun, 19 Jan 2014 14:22:13 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.599
X-Spam-Level: 
X-Spam-Status: No, score=-0.599 tagged_above=-999 required=5 tests=[BAYES_05=-0.5, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id rZm5_k824oQ8 for <aqm@ietfa.amsl.com>; Sun, 19 Jan 2014 14:22:11 -0800 (PST)
Received: from nm15-vm4.access.bullet.mail.bf1.yahoo.com (nm15-vm4.access.bullet.mail.bf1.yahoo.com [216.109.115.35]) by ietfa.amsl.com (Postfix) with ESMTP id 049351ADFA0 for <aqm@ietf.org>; Sun, 19 Jan 2014 14:22:10 -0800 (PST)
Received: from [66.196.81.157] by nm15.access.bullet.mail.bf1.yahoo.com with NNFMP; 19 Jan 2014 22:21:57 -0000
Received: from [98.139.244.53] by tm3.access.bullet.mail.bf1.yahoo.com with NNFMP; 19 Jan 2014 22:21:57 -0000
Received: from [127.0.0.1] by smtp115.sbc.mail.bf1.yahoo.com with NNFMP; 19 Jan 2014 22:21:57 -0000
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rogers.com; s=s1024; t=1390170117; bh=wKmpH0OpaUa2rPCcLs/Z2s2ES8M1jsH3vgI1BYItSZo=; h=X-Yahoo-Newman-Id:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:X-Rocket-Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:Subject:X-Enigmail-Version:Content-Type; b=sgGmr2ZtLGkMC5axwm9sqPkDbxDTTZawpTUau4Vp2fYk3Cx9FTNykPEZgQQfFOteXebtNMpdOf/0+sCq4Q3OzqBqY3EbIxg8lwzBxbVtvBBcyqQMoOAvIg/QNJz9Q28lqQbWqaSUdlu3NbLux+p7mZBuQsIX7+TpzGxC5kKm0Q4=
X-Yahoo-Newman-Id: 453526.71578.bm@smtp115.sbc.mail.bf1.yahoo.com
X-Yahoo-Newman-Property: ymail-3
X-YMail-OSG: bqInivYVM1nMRX8rqAnfBSHQshvzIZEwB7kG6RQ7W_p3MhN qGCRTbB7qCRfMl7OmdBSinGy4c0EZEqMz73Tc.668oW7L29aVi9UvOXYhQit 8GqvVjKTzdtVJ1Qelp3MIdmL.eNDHXOcZn38k0LAxJSxSMS6eqj59GqSDMBd DEYEeXv2X6X.uZB_ZY.SZUwORNEO.hVLrPTwH3I2vX4uhQuD0PrPhHKE8NXI JsN1MFDVNOY_eiJSQh7uP3YrBzrpZ3dHf1vDCL1aN5CqroAX4OcPjfgvplHQ J_AKgyFPPbqysFnCP_QzOUvgsPkps6Zw_8zWrBQuSX.LNXP6edsVtsaG4GGi jemK8VD3s460WUK7Eb73Q6PvS5zNJe58XIcRFQn.QKHTqhGSIlETDGpgMbmy M57ZhG497Suhd58rJYIKHxD.RSIrQXFmXl8nyuIB.DpydzHCvV5R7TyE0Lqj KK58jK5Wvy0orOmd4CiD5q9vfJx.TLnpxWzM9sAoL3dnZS1hCTL0UQOo3ZaA 5pJFblRupA_A56_gLwUc9li58fDJwho8qiZSUZL0fx6pQvmg.K5P05Q8mNjy uk5pkykhGT7Ke_Rh9QxdDduTHZF5lSH9i4xrrhjfjQV.683SHoJAdyuUb3Zm 9UquDDtIJ5KmWhb2mquNRwuljizk_IBJ5e3AklUNBs_UWGiu3L.nxDPeY
X-Yahoo-SMTP: sltvjZWswBCRD.ElTuB1l9j6s9wRYPpuyTNWOE5oEg--
X-Rocket-Received: from [192.168.0.10] (davec-b@99.225.150.135 with plain [98.139.221.125]) by smtp115.sbc.mail.bf1.yahoo.com with SMTP; 19 Jan 2014 14:21:57 -0800 PST
Message-ID: <52DC6001.3030501@rogers.com>
Date: Sun, 19 Jan 2014 18:30:09 -0500
From: David Collier-Brown <davec-b@rogers.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7
MIME-Version: 1.0
To: "aqm@ietf.org" <aqm@ietf.org>
X-Enigmail-Version: 1.6
Content-Type: multipart/alternative; boundary="------------070205090201080009080609"
Subject: [aqm] Comments on draft-ietf-aqm-recommendation-00, message 2 of several
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: davecb@spamcop.net
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 19 Jan 2014 22:22:14 -0000

This is a multi-part message in MIME format.
--------------070205090201080009080609
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

Section, paragraph (page)
=======  =========  ====
3.1 para 1 (p8)
   We call a flow "TCP-friendly" when it has a congestion response that
   approximates the average response expected of a TCP flow.  One
   example method of a TCP-friendly scheme is the TCP-Friendly Rate
   Control algorithm [RFC5348].  In this document, the term is used more
   generally to describe this and other algorithms that meet these
   goals.

I'd move this definition into section 3.1, as it's first paragraph. The
last paragraph of section enumerates the three different types, it's
legitimate to defer the definition until one paragraph later, and make
it easier to understand a TCP-friendly flow

I have a question about the phrase "TCP-friendly".  Responsive and
unresponsive are clearly applicable to bot TCP and UDP flows, which you
first discuss in section 3.2, just below.  It appears you're discussing
it as a general "congestion friendliness", of which RFC 5348 is an
example, and as if the reasoning can apply to UDP, which you first
discuss in the following section, 3.2

If this is correct, you may wish to not use TCP in the term, as it
appears to limit it to TCP, and not UDP implementation that have
improved congestion avoidance.

Regrettably, "congestion friendly" is not quite what we'd want (;-)) 
May I suggest "congestion-resistant"?

3.3 title (p9)
    Non-TCP-friendly Transport Protocols

Here I'd similarly suggest Non-Congestion-Resistant: they respond to
external signalling about congestion, but are not themselves resistant to it

3.3 para 2 (p9)
I'd split this into two paragraphs at  "Others may", so you have a
series of paragraphs giving example of non-congestion-resistant flows.
You might wish to make them numbered examples.

In the first sentence, you note that some may have poor implementations:
can you cite one, preferably a current one, but a historical blunder if
we don't have a current example.

after 3.3 (p9)

   The projected increase in the fraction of total Internet traffic for
   more aggressive flows in classes 2 and 3 clearly poses a threat to
   future Internet stability


[Niggle] Rather than classes 2 and 3, I'd recommend using their names,
 non-responsive and non-congestion-resistant.  We have several numbered
groups, and this one doesn't stand out, even though it's the most recent.

Do we have indications that these are becoming more prevalent?  One
example might be Google's increasing their initial congestion window: if
so, it might be mentioned here, not necessarily by company-name.

after 3.3, para 2 (p10)

    Another topic requiring consideration is the appropriate granularity
    of a "flow" when considering a queue management method.


If flow is defined in another RFC, we probably will wish to use their
definition
I've often seen the address and port-number quad referred to as a flow,
sometimes with further granularity as to time, which would be your case (2)



End of second day: comments, suggestions and brickbats cordially accepted!

--dave
[If anyone would prefer this as amendments to an editable format, I have
it in a doc file
at https://github.com/davecb/draft-ietf-aqm-recommendation-edits]

--------------070205090201080009080609
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<html>
  <head>
    <meta http-equiv="content-type" content="text/html;
      charset=ISO-8859-1">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <pre>Section, paragraph (page)
=======&nbsp; =========  ====
3.1 para 1 (p8)
   We call a flow "TCP-friendly" when it has a congestion response that
   approximates the average response expected of a TCP flow.  One
   example method of a TCP-friendly scheme is the TCP-Friendly Rate
   Control algorithm [RFC5348].  In this document, the term is used more
   generally to describe this and other algorithms that meet these
   goals.

</pre>
    I'd move this definition into section 3.1, as it's first paragraph.
    The last paragraph of section enumerates the three different types,
    it's legitimate to defer the definition until one paragraph later,
    and make it easier to understand a TCP-friendly flow<br>
    <br>
    I have a question about the phrase "TCP-friendly".&nbsp; Responsive and
    unresponsive are clearly applicable to bot TCP and UDP flows, which
    you first discuss in section 3.2, just below.&nbsp; It appears you're
    discussing it as a general "congestion friendliness", of which RFC
    5348 is an example, and as if the reasoning can apply to UDP, which
    you first discuss in the following section, 3.2<br>
    <br>
    If this is correct, you may wish to not use TCP in the term, as it
    appears to limit it to TCP, and not UDP implementation that have
    improved congestion avoidance.<br>
    <br>
    Regrettably, "congestion friendly" is not quite what we'd want
    (;-))&nbsp; May I suggest "congestion-resistant"?<br>
    <br>
    <pre>3.3 title (p9)
    Non-TCP-friendly Transport Protocols

</pre>
    Here I'd similarly suggest Non-Congestion-Resistant: they respond to
    external signalling about congestion, but are not themselves
    resistant to it<br>
    <br>
    3.3 para 2 (p9)<br>
    I'd split this into two paragraphs at&nbsp; "Others may", so you have a
    series of paragraphs giving example of non-congestion-resistant
    flows. You might wish to make them numbered examples.<br>
    <br>
    In the first sentence, you note that some may have poor
    implementations: can you cite one, preferably a current one, but a
    historical blunder if we don't have a current example.<br>
    <br>
    after 3.3 (p9)<br>
    <meta http-equiv="CONTENT-TYPE" content="text/html;
      charset=ISO-8859-1">
    <pre class="western" style="margin-bottom: 0in">   The projected increase in the fraction of total Internet traffic for
   more aggressive flows in classes 2 and 3 clearly poses a threat to
   future Internet stability


</pre>
    [Niggle] Rather than classes 2 and 3, I'd recommend using their
    names, &nbsp;non-responsive and non-congestion-resistant.&nbsp; We have
    several numbered groups, and this one doesn't stand out, even though
    it's the most recent.<br>
    <br>
    Do we have indications that these are becoming more prevalent?&nbsp; One
    example might be Google's increasing their initial congestion
    window: if so, it might be mentioned here, not necessarily by
    company-name.<br>
    <br>
    after 3.3, para 2 (p10)<br>
    <meta http-equiv="CONTENT-TYPE" content="text/html;
      charset=ISO-8859-1">
    <p class="western" style="margin-bottom: 0in">&nbsp;&nbsp;&nbsp; Another topic
      requiring
      consideration is the appropriate granularity<br>
      &nbsp; &nbsp; of a "flow" when considering a queue management method.</p>
    <title></title>
    <meta name="GENERATOR" content="LibreOffice 3.5 (Linux)">
    <style type="text/css">
	<!--
		@page { margin: 0.79in }
		PRE { margin-left: 0.1in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		PRE.western { so-language: en-US }
		PRE.cjk { font-family: "WenQuanYi Zen Hei Sharp", monospace; so-language: zh-CN }
		PRE.ctl { font-family: "Lohit Devanagari", monospace; so-language: hi-IN }
		P { margin-bottom: 0.08in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		P.western { font-family: "Times New Roman", serif; font-size: 12pt; so-language: en-US }
		P.cjk { font-family: "WenQuanYi Zen Hei Sharp"; font-size: 12pt; so-language: zh-CN }
		P.ctl { font-family: "Lohit Devanagari"; font-size: 12pt; so-language: hi-IN }
	-->
	</style><br>
    <span class="st"> If flow is defined in another RFC, we probably
      will wish to use </span>their definition<br>
    I've often seen the address and port-number quad referred to as a
    flow, <span class="st">sometimes with further granularity as to
      time, which would be your case (2)</span><br>
    <br>
    <br>
    <br>
    End of second day: comments, suggestions and brickbats cordially
    accepted!<br>
    <br>
    --dave<br>
    [If anyone would prefer this as amendments to an editable format, I
    have it in a doc file<br>
    at <a class="moz-txt-link-freetext" href="https://github.com/davecb/draft-ietf-aqm-recommendation-edits">https://github.com/davecb/draft-ietf-aqm-recommendation-edits</a>]<br>
  </body>
</html>

--------------070205090201080009080609--

From touch@isi.edu  Mon Jan 20 11:17:56 2014
Return-Path: <touch@isi.edu>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9B6F51A0216 for <aqm@ietfa.amsl.com>; Mon, 20 Jan 2014 11:17:56 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.435
X-Spam-Level: 
X-Spam-Status: No, score=-2.435 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RP_MATCHES_RCVD=-0.535] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EFfG27vb9H7R for <aqm@ietfa.amsl.com>; Mon, 20 Jan 2014 11:17:54 -0800 (PST)
Received: from darkstar.isi.edu (darkstar.isi.edu [128.9.128.127]) by ietfa.amsl.com (Postfix) with ESMTP id 269321A01DC for <aqm@ietf.org>; Mon, 20 Jan 2014 11:17:54 -0800 (PST)
Received: from [192.168.1.91] (pool-71-105-87-112.lsanca.dsl-w.verizon.net [71.105.87.112]) (authenticated bits=0) by darkstar.isi.edu (8.13.8/8.13.8) with ESMTP id s0KJHT4T015886 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Mon, 20 Jan 2014 11:17:33 -0800 (PST)
Mime-Version: 1.0 (Mac OS X Mail 7.1 \(1827\))
Content-Type: text/plain; charset=iso-8859-1
From: Joe Touch <touch@isi.edu>
In-Reply-To: <52DC6001.3030501@rogers.com>
Date: Mon, 20 Jan 2014 11:17:31 -0800
Content-Transfer-Encoding: quoted-printable
Message-Id: <3B61AD42-D51F-4977-8F33-3344DED65691@isi.edu>
References: <52DC6001.3030501@rogers.com>
To: davecb@spamcop.net
X-Mailer: Apple Mail (2.1827)
X-ISI-4-43-8-MailScanner: Found to be clean
X-MailScanner-From: touch@isi.edu
Cc: "aqm@ietf.org" <aqm@ietf.org>
Subject: Re: [aqm] Comments on draft-ietf-aqm-recommendation-00, message 2 of several
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 20 Jan 2014 19:17:56 -0000

On Jan 19, 2014, at 3:30 PM, David Collier-Brown <davec-b@rogers.com> =
wrote:

> Section, paragraph (page)
> =3D=3D=3D=3D=3D=3D=3D  =3D=3D=3D=3D=3D=3D=3D=3D=3D  =3D=3D=3D=3D
> 3.1 para 1 (p8)
>    We call a flow "TCP-friendly" when it has a congestion response =
that
>    approximates the average response expected of a TCP flow.  One
>    example method of a TCP-friendly scheme is the TCP-Friendly Rate
>    Control algorithm [RFC5348].  In this document, the term is used =
more
>    generally to describe this and other algorithms that meet these
>    goals.
>=20
>=20
> I'd move this definition into section 3.1, as it's first paragraph. =
The last paragraph of section enumerates the three different types, it's =
legitimate to defer the definition until one paragraph later, and make =
it easier to understand a TCP-friendly flow
>=20
> I have a question about the phrase "TCP-friendly".  Responsive and =
unresponsive are clearly applicable to bot TCP and UDP flows, which you =
first discuss in section 3.2, just below.  It appears you're discussing =
it as a general "congestion friendliness", of which RFC 5348 is an =
example, and as if the reasoning can apply to UDP, which you first =
discuss in the following section, 3.2
>=20
> If this is correct, you may wish to not use TCP in the term, as it =
appears to limit it to TCP, and not UDP implementation that have =
improved congestion avoidance.

IMO, "responsive" isn't correct.

TCP-friendly also means friendly to TCP, i.e., there are numerous =
responsive variants of TCP congestion control that are not particularly =
friendly to the current recommended standard.

Typically, when we mean congestion responsive we say that; when we say =
"TCP friendly", we mean that it gets no more than a similar proportion =
of resources as the current standard TCP when competing for resources.

> Regrettably, "congestion friendly" is not quite what we'd want (;-))  =
May I suggest "congestion-resistant"?

I don't think that's what's sought here, ultimately. There are a nearly =
infinite number of congestion resistant solutions that are not practical =
for use in the current Internet because of how they interact with the =
current TCP standards.

Joe=

From davec-b@rogers.com  Mon Jan 20 12:34:27 2014
Return-Path: <davec-b@rogers.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BD8AC1A024A for <aqm@ietfa.amsl.com>; Mon, 20 Jan 2014 12:34:27 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GpTbfA4J1ZtY for <aqm@ietfa.amsl.com>; Mon, 20 Jan 2014 12:34:25 -0800 (PST)
Received: from nm26-vm4.access.bullet.mail.gq1.yahoo.com (nm26-vm4.access.bullet.mail.gq1.yahoo.com [216.39.63.114]) by ietfa.amsl.com (Postfix) with ESMTP id AFA721A0169 for <aqm@ietf.org>; Mon, 20 Jan 2014 12:34:24 -0800 (PST)
Received: from [216.39.60.167] by nm26.access.bullet.mail.gq1.yahoo.com with NNFMP; 20 Jan 2014 20:34:25 -0000
Received: from [67.195.22.117] by tm3.access.bullet.mail.gq1.yahoo.com with NNFMP; 20 Jan 2014 20:34:24 -0000
Received: from [127.0.0.1] by smtp112.sbc.mail.gq1.yahoo.com with NNFMP; 20 Jan 2014 20:34:24 -0000
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rogers.com; s=s1024; t=1390250064; bh=jVtl1G+ZeVPrikeejClIY8K8EWIT/grxhgxLz9Gm94c=; h=X-Yahoo-Newman-Id:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:X-Rocket-Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:Subject:References:In-Reply-To:X-Enigmail-Version:Content-Type:Content-Transfer-Encoding; b=aX9+TcfrzW2gF7RnPQVLQRQhjnL8mHVGh4a+aEoJpc3zxCDdXrqfnz7y+c70ntfL//L1reB0cTMggvLoZTIz3DT3QY5yHJNc8Cjk1Yc/5FglwfNnhv0uPdnjntap75I6BNtxWpaPwgDwYrLQbFJoqJOCM34w+8diMeAQwwGZ6TQ=
X-Yahoo-Newman-Id: 868124.70328.bm@smtp112.sbc.mail.gq1.yahoo.com
X-Yahoo-Newman-Property: ymail-3
X-YMail-OSG: NAc9z80VM1nxbvhNfMMVlvF.5S93eEDfPNVJB2kKO5.IfJD iarBYX3LRqm.Pci56zu3UPw5Rqvo0bS8MZ4oOrm.Ev7hdeMFJ6CTvpiaC2qW CElUlXmf25sJ_.edznxZ.yDurmzIJfli58wkPD0pjUXvMOIK4iwuSuleDE.V HkFxLaopcuNjc4BkGUtCI2vBWjmUBta6PEah3S5XzE_sI.14gPmPunNsPIFA C4Om0mxN0Wkg8zp8Fj_dUhy7SvR0_VwDeFryVr8YASIDFTM7Zr.TRgUW41fF YRNA.KevXXU0L14p_DXk2K1wOWb16YBCe8qPLjYBViPOuJmpTYX.8UrBMkE6 de15jlM9Dee6UxkNelauM9S_hGNMePsIl1BngbJBKm6vDKRBGnk7oawwMDi0 8kyP_cnbhyC5clLqlyUfrQnkckJcRx3mk7W6uY2jl9OgXaDhnUqx4IlrP5ju ZPXwgK17A5MHJ2hwPm_74hBaexpGWGd2jB8V.CrMrTrykeauH.kRxaJBMtYz J8fA30lWYPxndOZjhPoBJ50FlaMNhat5YQRCB1SgwrEL1wBKx8pL8cz9zCyP Z6gJ_9CI-
X-Yahoo-SMTP: sltvjZWswBCRD.ElTuB1l9j6s9wRYPpuyTNWOE5oEg--
X-Rocket-Received: from [10.111.100.153] (davec-b@74.213.188.46 with plain [67.195.22.104]) by smtp112.sbc.mail.gq1.yahoo.com with SMTP; 20 Jan 2014 12:34:24 -0800 PST
Message-ID: <52DD884F.5010200@rogers.com>
Date: Mon, 20 Jan 2014 15:34:23 -0500
From: David Collier-Brown <davec-b@rogers.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7
MIME-Version: 1.0
To: aqm@ietf.org
References: <52DC6001.3030501@rogers.com> <3B61AD42-D51F-4977-8F33-3344DED65691@isi.edu>
In-Reply-To: <3B61AD42-D51F-4977-8F33-3344DED65691@isi.edu>
X-Enigmail-Version: 1.6
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Subject: Re: [aqm] Comments on draft-ietf-aqm-recommendation-00, message 2 of several
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: davecb@spamcop.net
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 20 Jan 2014 20:34:27 -0000

On 01/20/2014 02:17 PM, Joe Touch wrote:
> On Jan 19, 2014, at 3:30 PM, David Collier-Brown <davec-b@rogers.com> wrote:
>
>> Section, paragraph (page)
>> =======  =========  ====
>> 3.1 para 1 (p8)
>>    We call a flow "TCP-friendly" when it has a congestion response that
>>    approximates the average response expected of a TCP flow.  One
>>    example method of a TCP-friendly scheme is the TCP-Friendly Rate
>>    Control algorithm [RFC5348].  In this document, the term is used more
>>    generally to describe this and other algorithms that meet these
>>    goals.
>>
>>
>> I'd move this definition into section 3.1, as it's first paragraph. The last paragraph of section enumerates the three different types, it's legitimate to defer the definition until one paragraph later, and make it easier to understand a TCP-friendly flow
>>
>> I have a question about the phrase "TCP-friendly".  Responsive and unresponsive are clearly applicable to bot TCP and UDP flows, which you first discuss in section 3.2, just below.  It appears you're discussing it as a general "congestion friendliness", of which RFC 5348 is an example, and as if the reasoning can apply to UDP, which you first discuss in the following section, 3.2
>>
>> If this is correct, you may wish to not use TCP in the term, as it appears to limit it to TCP, and not UDP implementation that have improved congestion avoidance.
> IMO, "responsive" isn't correct.
>
> TCP-friendly also means friendly to TCP, i.e., there are numerous responsive variants of TCP congestion control that are not particularly friendly to the current recommended standard.
>
> Typically, when we mean congestion responsive we say that; when we say "TCP friendly", we mean that it gets no more than a similar proportion of resources as the current standard TCP when competing for resources.
>
>> Regrettably, "congestion friendly" is not quite what we'd want (;-))  May I suggest "congestion-resistant"?
> I don't think that's what's sought here, ultimately. There are a nearly infinite number of congestion resistant solutions that are not practical for use in the current Internet because of how they interact with the current TCP standards.
>
> Joe
> _______________________________________________
> aqm mailing list
> aqm@ietf.org
> https://www.ietf.org/mailman/listinfo/aqm
>
Excellent, thanks!

--dave

-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
davecb@spamcop.net           |                      -- Mark Twain
(416) 223-8968


From Ruediger.Geib@telekom.de  Mon Jan 20 23:38:04 2014
Return-Path: <Ruediger.Geib@telekom.de>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EA1F31A005E; Mon, 20 Jan 2014 23:38:04 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.886
X-Spam-Level: 
X-Spam-Status: No, score=-0.886 tagged_above=-999 required=5 tests=[BAYES_20=-0.001, HELO_EQ_DE=0.35, RCVD_IN_DNSWL_LOW=-0.7, RP_MATCHES_RCVD=-0.535] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id AsUD4zxMU3uX; Mon, 20 Jan 2014 23:38:02 -0800 (PST)
Received: from tcmail23.telekom.de (tcmail23.telekom.de [80.149.113.243]) by ietfa.amsl.com (Postfix) with ESMTP id F1B491A0058; Mon, 20 Jan 2014 23:38:01 -0800 (PST)
Received: from he113657.emea1.cds.t-internal.com ([10.134.99.17]) by tcmail21.telekom.de with ESMTP/TLS/AES128-SHA; 21 Jan 2014 08:38:00 +0100
Received: from HE111643.EMEA1.CDS.T-INTERNAL.COM ([10.134.93.12]) by HE113657.emea1.cds.t-internal.com ([::1]) with mapi; Tue, 21 Jan 2014 08:38:00 +0100
From: <Ruediger.Geib@telekom.de>
To: <bob.briscoe@bt.com>
Date: Tue, 21 Jan 2014 08:37:59 +0100
Thread-Topic: [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or tunnel protocols?
Thread-Index: Ac7ZqeZ/7AQswgk0TByj6KTyywQpeg80Potg
Message-ID: <CA7A7C64CC4ADB458B74477EA99DF6F502347353E1@HE111643.EMEA1.CDS.T-INTERNAL.COM>
References: <201311042203.rA4M3lo0026458@bagheera.jungle.bt.co.uk>
In-Reply-To: <201311042203.rA4M3lo0026458@bagheera.jungle.bt.co.uk>
Accept-Language: en-US, de-DE
Content-Language: de-DE
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US, de-DE
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Mailman-Approved-At: Tue, 21 Jan 2014 00:29:01 -0800
Cc: tsvwg@ietf.org, aqm@ietf.org
Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or tunnel protocols?
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 21 Jan 2014 07:38:05 -0000

Hi Bob,

I support the issue being picked up by IETF. What can be done within the bo=
unds of IETF responsibility should be done. If ECN is seeing deployment, es=
pecially ECN support for IP over VLAN over IP/MPLS may be of interest. Furt=
her, ECN over LTE radio Access may be relevant (but my expertise is too lim=
ited to judge details).

Regards,

Ruediger =20

-----Urspr=FCngliche Nachricht-----
Von: tsvwg-bounces@ietf.org [mailto:tsvwg-bounces@ietf.org] Im Auftrag von =
Bob Briscoe
Gesendet: Montag, 4. November 2013 23:04
An: tsvwg IETF list; AQM IETF list
Cc: draft-briscoe-tsvwg-ecn-encap-guidelines@tools.ietf.org
Betreff: [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or tunnel =
protocols?

Folks,

Pls respond if you support this being adopted as a work-group item in the I=
ETF transport services w-g (tsvwg). The WG chairs need visibility of intere=
st.
Even better, if you're willing to read / comment / review / implement

Guidelines for Adding Congestion Notification to Protocols that Encapsulate=
 IP <http://tools.ietf.org/html/draft-briscoe-tsvwg-ecn-encap-guidelines>

Abstract

    The purpose of this document is to guide the design of congestion
    notification in any lower layer or tunnelling protocol that
    encapsulates IP.  The aim is for explicit congestion signals to
    propagate consistently from lower layer protocols into IP.  Then the
    IP internetwork layer can act as a portability layer to carry
    congestion notification from non-IP-aware congested nodes up to the
    transport layer (L4).  Following these guidelines should assure
    interworking between new lower layer congestion notification
    mechanisms, whether specified by the IETF or other standards bodies.


[Cross-posting tsvwg & aqm, just in case]


Bob Briscoe,
also for co-authors Pat Thaler and John Kaippallimalil


________________________________________________________________
Bob Briscoe,                                                  BT=20


From touch@isi.edu  Tue Jan 21 09:36:42 2014
Return-Path: <touch@isi.edu>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 174C51A042A for <aqm@ietfa.amsl.com>; Tue, 21 Jan 2014 09:36:42 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.735
X-Spam-Level: 
X-Spam-Status: No, score=-4.735 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.535] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id y1vkDpQeUWRt for <aqm@ietfa.amsl.com>; Tue, 21 Jan 2014 09:36:38 -0800 (PST)
Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by ietfa.amsl.com (Postfix) with ESMTP id D73551A018E for <aqm@ietf.org>; Tue, 21 Jan 2014 09:36:38 -0800 (PST)
Received: from [128.9.160.166] (abc.isi.edu [128.9.160.166]) (authenticated bits=0) by boreas.isi.edu (8.13.8/8.13.8) with ESMTP id s0LHZgLS002677 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Tue, 21 Jan 2014 09:35:42 -0800 (PST)
Message-ID: <52DEAFEE.9070803@isi.edu>
Date: Tue, 21 Jan 2014 09:35:42 -0800
From: Joe Touch <touch@isi.edu>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:24.0) Gecko/20100101 Thunderbird/24.2.0
MIME-Version: 1.0
To: davecb@spamcop.net, "aqm@ietf.org" <aqm@ietf.org>
References: <52DB2E1F.50009@rogers.com>
In-Reply-To: <52DB2E1F.50009@rogers.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-ISI-4-43-8-MailScanner: Found to be clean
X-MailScanner-From: touch@isi.edu
Subject: Re: [aqm] Comments on draft-ietf-aqm-recommendation-00, message 1 of several
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 21 Jan 2014 17:36:42 -0000

On 1/18/2014 5:45 PM, David Collier-Brown wrote:
> Section, paragraph (page)
> =======  =========  ====
>
> 1, 3 (p3)
>
>     ... mechanisms [RFC5681], while necessary and powerful, are not
>     sufficient to provide good service in all circumstances.  Basically,
>     there is a limit to how much control can be accomplished from the
>     edges of the network.  Some mechanisms are needed in the network
>     devices to complement the endpoint congestion avoidance mechanisms.
>
> This reads a bit like middleboxes aren't using anything, and that they
> need different
> mechanisms.  I'd suggest they need to be as active as using the same
> control
> mechanisms as the endpoints, or at least their components of the control
> mechanisms.

That might not be a good idea, FWIW.

Feedback control systems are known to interact in complex ways. Chaining 
a series of the same congestion control mechanisms, and embedding them 
in another E2E congestion control mechanism is a recipe for unintended 
consequences.

> How about "Network devices need to actively use the same mechanisms to
> provide hop-by-hop congestion avoidance as are used by the endpoints,
> especially the newer mechanisms."

See above.

A key point about the E2E argument is that hop-based mechanisms cannot 
necessarily be composed to provide an equivalent E2E service.

> That reflects the fact that the
> experimentation is going on at or quite near the edge, and the middle
> has been a bit left out.

Agreed, because another point of the E2E argumetn is that such hop-based 
mechanisms can be critical for E2E performance.

So while I agree that network devices need to engage some sort of 
control mechanisms, it's not at all clear whether they should be 
identical to the E2E mechanisms.

I would not be surprised if hop mechanisms didn't have to operate over a 
completely different timescale, for example.

Joe

> I'd also suggest the second sentence would be stronger if it doesn't
> begin with "basically".
>
>
> 1, 4 (p3)
>     ... To a rough approximation, queue management algorithms
>     manage the length of packet queues by marking or dropping packets
>     when necessary or appropriate, while scheduling algorithms determine ...
>
> Based on the description in 1.1.2, para 2 (the one about queues I
> mentioned in the prefatory remarks), I suggest we say here that queue
> management algorithms "/minimize/ the length of packet queues".
>
> This is a bit of a truism in the queuing world, and a different mind-set
> than buffer management. One wouldn't say "minimize buffering", as that
> would draw an entirely backwards picture in the reader's mind (;-)), but
> we do wish to minimize the waiting-in-queue, for both throughput and
> latency reasons, not just for one or the other.
>
>
> 1, 6 (p3)
>     The second issue, discussed in Section 3 of this memo, is the
>     potential for future congestive collapse of the Internet due to
>     ...
>     available.  It is imperative that this work be energetically pursued,
>     to ensure the future stability of the Internet.
>
> Unless it is impossible for congestive collapse to happen now, I'd
> suggest putting this into the present tense: say "potential for
> congestive collapse" and "retain the stability of the internet"
> As it stands, it sounds like we're arguing that this can only happen at
> some future date, when something worsens, and so the reader could
> reasonably conclude that they need not act.
>
> 1, 8 (p4)
>     The discussion in this memo applies to "best-effort" traffic, which
>     is to say, traffic generated by applications that accept the
>     occasional loss, duplication, or reordering of traffic in flight.  It
>     also applies to other traffic, such as real-time traffic that can
>     adapt its sending rate to reduce loss and/or delay.  It is most
>     effective, when the adaption occurs on time scales of a single RTT or
>     a small number of RTTs, for elastic traffic [RFC1633].
>
> I find this just plain hard to read.  I'm going to assume you mean to
> identify the traffic regimes we consider applicable, and in indicate
> their desirability. If so, may I suggest:
>
>     The discussion in this memo applies to most observed kinds of traffic, but not to <specific
>     technical term goes here> ones, where no delay, loss or reordering can be permitted.  It works
>     for so-called "real-time" traffic that can adapt its sending rate to reduce loss and/or delay,
>     and is optimal for the common case, "best effort" traffic, traffic that can accept the
>     occasional loss, duplication, or reordering of traffic in flight. It is most effective for
>     <a specific technical term goes here too?> traffic where such events occur on time scales of a
>     single RTT or a small number of RTTs.
>
>
> I initially through you were referring forward to the responsive,
> unresponsive and tcp-friendly flows you mention in section 3, but they
> clearly don't fit: can you flesh this paragraph out with the kinds of
> traffic you mean, or indicate if you meant something else entirely?
>
> 2, 1 (p4)
>
>     The traditional technique for managing the queue length in a network
>     device is to set a maximum length (in terms of packets) for each
>     queue, accept packets for the queue until the maximum length is
>     reached, then reject (drop) subsequent incoming packets until the
>     queue decreases because a packet from the queue has been transmitted.
>
> I'd break the paragraph here and let its name be a separate paragraph.
>
> I'd follow it with a statement of why we have to deal with queues, such as
>
>      These queues build up wherever there is a connection between an incoming and an
>      outgoing stream of packets, and a buffer is used to hold packets where they are
>      being routed, where the input and output are running at different speeds, or
>      where a group of packets arrive together at the input and they need to be held until
>      output processing can drain them away.
>
> This then distinguishes between the three uses of the buffer, and
> motivates the following argument.
>
> 2.1, 1 (p4)
>
>         In some situations tail drop allows a single connection or a few
>         flows to monopolize queue space, preventing other connections
>         from getting room in the queue.  This "lock-out" phenomenon is
>         often the result of synchronization or other timing effects.
>
>
> I think it might be better to say "inadvertent synchronization", as that
> avoids suggesting that some separate and intentional synchronization was
> causing the problem.
>
> 2.1, 2 (p4)
>
>         The tail drop discipline allows queues to maintain a full (or,
>         almost full) buffer for long periods of time, since tail drop
>         signals congestion (via a packet drop) only when the queue has
>         become full.
>
> I changed "status" to "buffer" in the above. Following this sentence, may I
> suggest we skip immediately to
>
>         The point of buffering in the network is to absorb data bursts
>         and to transmit them during the (hopefully) ensuing bursts of
>         silence.  This is essential to permit the transmission of bursty
>         data.  Normally small queues are preferred in network devices,
>         with sufficient queue capacity to absorb the bursts.
>
> then
>
>         Even though TCP constrains the congestion window of
>         a flow, packets often arrive at network devices in bursts
>         [Leland94].  If the queue is full or almost full, an arriving
>         burst will cause multiple packets to be dropped.  This can result
>         in a global synchronization of flows throttling back, followed by
>         a sustained period of lowered link utilization, reducing overall
>         throughput.
>
>         The counter-intuitive result is that maintaining normally-small
>         buffer residencies (queues) can result in higher throughput as
>         well as lower end-to-end delay.  In summary, buffer/queue limits
>         should not reflect the steady state queues we want to be maintained
>         in the network; instead, they should reflect the size of bursts that
>         a network device needs to absorb.
>
> I Inserted "buffer residencies (queues)" in place of "queues" and
> "buffer/queue limits" in place of "queue limits", then we could follow
> up with
>
>         The naive assumption might be that there is a simple tradeoff
>         between delay and throughput, and that the recommendation that
>         queues be maintained in a "non-full" state essentially translates
>         to a recommendation that low end-to-end delay is more important
>         than high throughput.
>
> I'd then conclude with
>
>          As we see above, this is by no means the case. Small queues and
>          normally-empty buffers maximize throughput as well.
>
>
> 2.2, 5 (p5)
>
>     We know in general how to solve the full-queues problem for
>     "responsive" flows, i.e., those flows that throttle back in response
>     to congestion notification.  In the current Internet, dropped packets
>     provide a critical mechanism indicating congestion notification to
>     hosts.  The solution to the full-queues problem is for network
>     devices to drop packets before a queue becomes full, so that hosts
>     can respond to congestion before buffers overflow.  We call such a
>     proactive approach AQM.  By dropping packets before buffers overflow,
>     AQM allows network devices to control when and how many packets to
>     drop.
>
> As we're going to use ECN fairly soon as part of AQM, we need to
> foreshadow that, and this is also a good place to define AQM.  I suggest
> changing  "The solution to the full-queues problem ... We call such a
> proactive approach AQM. " to
>
>     The solution to the full-queues problem is for network devices to drop packets
>     or otherwise notify the sender before a queue becomes full, so that hosts
>     can respond to congestion before buffers overflow.  We define such a proactive
>     approach to be/Advanced Queue Management/  (AQM).  By dropping or marking packets
>     before buffers overflow, AQM allows network devices to maximize performance.
>
> End of first day: comments, suggestions and brickbats cordially accepted!
>
> --dave
> [If anyone would prefer this as amendments to an editable format, just say]
>
>
> _______________________________________________
> aqm mailing list
> aqm@ietf.org
> https://www.ietf.org/mailman/listinfo/aqm
>

From ingemar.s.johansson@ericsson.com  Tue Jan 21 12:29:12 2014
Return-Path: <ingemar.s.johansson@ericsson.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A6B191A02E7; Tue, 21 Jan 2014 12:29:12 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.851
X-Spam-Level: 
X-Spam-Status: No, score=-3.851 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HELO_EQ_SE=0.35, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id lMQqFR-F-eoZ; Tue, 21 Jan 2014 12:29:09 -0800 (PST)
Received: from mailgw1.ericsson.se (mailgw1.ericsson.se [193.180.251.45]) by ietfa.amsl.com (Postfix) with ESMTP id 260031A02DC; Tue, 21 Jan 2014 12:29:08 -0800 (PST)
X-AuditID: c1b4fb2d-b7f5d8e000002a7b-5a-52ded89468a5
Received: from ESESSHC009.ericsson.se (Unknown_Domain [153.88.253.124]) by mailgw1.ericsson.se (Symantec Mail Security) with SMTP id A5.2F.10875.498DED25; Tue, 21 Jan 2014 21:29:08 +0100 (CET)
Received: from ESESSMB205.ericsson.se ([169.254.5.205]) by ESESSHC009.ericsson.se ([153.88.183.45]) with mapi id 14.02.0387.000; Tue, 21 Jan 2014 21:29:07 +0100
From: Ingemar Johansson S <ingemar.s.johansson@ericsson.com>
To: "Ruediger.Geib@telekom.de" <Ruediger.Geib@telekom.de>, "bob.briscoe@bt.com" <bob.briscoe@bt.com>
Thread-Topic: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or tunnel protocols?
Thread-Index: AQHPFuN4cehttPFQ50yooU+SLXqCYZqPn5Ow
Date: Tue, 21 Jan 2014 20:29:06 +0000
Message-ID: <81564C0D7D4D2A4B9A86C8C7404A13DA31EC250B@ESESSMB205.ericsson.se>
References: <201311042203.rA4M3lo0026458@bagheera.jungle.bt.co.uk> <CA7A7C64CC4ADB458B74477EA99DF6F502347353E1@HE111643.EMEA1.CDS.T-INTERNAL.COM>
In-Reply-To: <CA7A7C64CC4ADB458B74477EA99DF6F502347353E1@HE111643.EMEA1.CDS.T-INTERNAL.COM>
Accept-Language: sv-SE, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [153.88.183.150]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFrrOLMWRmVeSWpSXmKPExsUyM+Jvje6UG/eCDHo3C1ms2SdpMX39F0aL D9M4LI69ucvmwOLR9mUyk8eSJT+ZPNpeKgQwR3HZpKTmZJalFunbJXBltEx7yFjwXqTi6ut3 zA2MxwW6GDk4JARMJC6tV+1i5AQyxSQu3FvP1sXIxSEkcIhR4sW9V8wQzhJGiSu3W5hAqtgE bCRWHvrOCGKLCKRJnP54hx1kELOAi8TU/cwgYWGBRInza88yQZQkSRy6tQGq3Eji2N7ZYDUs AqoS854cBqvhFfCVWHxpD9SuyYwS95Y/BJvJKRAtMf2LOkgNo4CsxP3v91hAbGYBcYlbT+Yz QRwtILFkz3lmCFtU4uXjf6wQtpLEiu2XGCHq9SRuTJ3CBmFrSyxb+JoZYq+gxMmZT1gmMIrN QjJ2FpKWWUhaZiFpWcDIsoqRPTcxMye93HATIzByDm75rbuD8dQ5kUOM0hwsSuK8H946BwkJ pCeWpGanphakFsUXleakFh9iZOLglGpgVPx7eprUprLpu6KL/Ge1KG7++Pf81O+6BV/WXtKQ vSwxVdHm/h4mJ7O6lWWH43/ble6JizrkzdzvqCMhESZ8T1rrvWzbh6T4HaXhS69lWjyf23a/ Zrmd/qH7fwsKCpcV6AVs3DBNYt6yI4XZt9b0K/g4Je3d5Oj42fdE/IH4OX0Ln7owZhW6KLEU ZyQaajEXFScCABmYLVhqAgAA
Cc: "tsvwg@ietf.org" <tsvwg@ietf.org>, "aqm@ietf.org" <aqm@ietf.org>
Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or tunnel protocols?
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 21 Jan 2014 20:29:12 -0000

Hi

Please note that ECN over LTE radio access is already standardized in 3GPP =
TS 36.300 (see 11.6 in http://www.3gpp.org/ftp/specs/archive/36_series/36.3=
00/36300-c00.zip )

/Ingemar
> -----Original Message-----
> From: Ruediger.Geib@telekom.de [mailto:Ruediger.Geib@telekom.de]
> Sent: den 21 januari 2014 08:38
> To: bob.briscoe@bt.com
> Cc: tsvwg@ietf.org; aqm@ietf.org
> Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to L=
2
> or tunnel protocols?
>=20
> Hi Bob,
>=20
> I support the issue being picked up by IETF. What can be done within the
> bounds of IETF responsibility should be done. If ECN is seeing deployment=
,
> especially ECN support for IP over VLAN over IP/MPLS may be of interest.
> Further, ECN over LTE radio Access may be relevant (but my expertise is t=
oo
> limited to judge details).
>=20
> Regards,
>=20
> Ruediger
>=20
> -----Urspr=FCngliche Nachricht-----
> Von: tsvwg-bounces@ietf.org [mailto:tsvwg-bounces@ietf.org] Im Auftrag
> von Bob Briscoe
> Gesendet: Montag, 4. November 2013 23:04
> An: tsvwg IETF list; AQM IETF list
> Cc: draft-briscoe-tsvwg-ecn-encap-guidelines@tools.ietf.org
> Betreff: [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or tunne=
l
> protocols?
>=20
> Folks,
>=20
> Pls respond if you support this being adopted as a work-group item in the
> IETF transport services w-g (tsvwg). The WG chairs need visibility of int=
erest.
> Even better, if you're willing to read / comment / review / implement
>=20
> Guidelines for Adding Congestion Notification to Protocols that Encapsula=
te
> IP <http://tools.ietf.org/html/draft-briscoe-tsvwg-ecn-encap-guidelines>
>=20
> Abstract
>=20
>     The purpose of this document is to guide the design of congestion
>     notification in any lower layer or tunnelling protocol that
>     encapsulates IP.  The aim is for explicit congestion signals to
>     propagate consistently from lower layer protocols into IP.  Then the
>     IP internetwork layer can act as a portability layer to carry
>     congestion notification from non-IP-aware congested nodes up to the
>     transport layer (L4).  Following these guidelines should assure
>     interworking between new lower layer congestion notification
>     mechanisms, whether specified by the IETF or other standards bodies.
>=20
>=20
> [Cross-posting tsvwg & aqm, just in case]
>=20
>=20
> Bob Briscoe,
> also for co-authors Pat Thaler and John Kaippallimalil
>=20
>=20
> __________________________________________________________
> ______
> Bob Briscoe,                                                  BT
>=20


From bob.briscoe@bt.com  Tue Jan 21 14:49:48 2014
Return-Path: <bob.briscoe@bt.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D25411A0343; Tue, 21 Jan 2014 14:49:48 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.436
X-Spam-Level: 
X-Spam-Status: No, score=-2.436 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RP_MATCHES_RCVD=-0.535, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tnczsPyfm-0s; Tue, 21 Jan 2014 14:49:45 -0800 (PST)
Received: from hubrelay-rd.bt.com (hubrelay-rd.bt.com [62.239.224.98]) by ietfa.amsl.com (Postfix) with ESMTP id 662A81A0256; Tue, 21 Jan 2014 14:49:44 -0800 (PST)
Received: from EVMHR71-UKRD.domain1.systemhost.net (10.36.3.109) by EVMHR66-UKRD.bt.com (10.187.101.21) with Microsoft SMTP Server (TLS) id 14.3.158.1; Tue, 21 Jan 2014 22:49:43 +0000
Received: from EPHR01-UKIP.domain1.systemhost.net (147.149.196.177) by EVMHR71-UKRD.domain1.systemhost.net (10.36.3.109) with Microsoft SMTP Server (TLS) id 8.3.279.1; Tue, 21 Jan 2014 22:49:43 +0000
Received: from bagheera.jungle.bt.co.uk (132.146.168.158) by EPHR01-UKIP.domain1.systemhost.net (147.149.196.177) with Microsoft SMTP Server id 14.2.347.0; Tue, 21 Jan 2014 22:49:41 +0000
Received: from BTP075694.jungle.bt.co.uk ([10.111.123.72])	by bagheera.jungle.bt.co.uk (8.13.5/8.12.8) with ESMTP id s0LMneWw026346;	Tue, 21 Jan 2014 22:49:41 GMT
Message-ID: <201401212249.s0LMneWw026346@bagheera.jungle.bt.co.uk>
X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9
Date: Tue, 21 Jan 2014 22:49:35 +0000
To: Ingemar Johansson S <ingemar.s.johansson@ericsson.com>, "Ruediger.Geib@telekom.de" <Ruediger.Geib@telekom.de>
From: Bob Briscoe <bob.briscoe@bt.com>
In-Reply-To: <81564C0D7D4D2A4B9A86C8C7404A13DA31EC250B@ESESSMB205.ericss on.se>
References: <201311042203.rA4M3lo0026458@bagheera.jungle.bt.co.uk> <CA7A7C64CC4ADB458B74477EA99DF6F502347353E1@HE111643.EMEA1.CDS.T-INTERNAL.COM> <81564C0D7D4D2A4B9A86C8C7404A13DA31EC250B@ESESSMB205.ericsson.se>
MIME-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"; format=flowed
Content-Transfer-Encoding: quoted-printable
X-Scanned-By: MIMEDefang 2.56 on 132.146.168.158
Cc: "tsvwg@ietf.org" <tsvwg@ietf.org>, "aqm@ietf.org" <aqm@ietf.org>
Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or tunnel protocols?
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 21 Jan 2014 22:49:49 -0000

Ingemar,

1) Thx for the pointer. We should add this as=20
another example where a lower layer marks the IP=20
header (the example we already have is the L3 switch in the Ethernet world).

2) It doesn't talk about propagation of ECN=20
markings during encap & decap, which is one of=20
the things ecn-encap-guidelines aims to give=20
guidelines on. Do you think a doc like the draft=20
we've done would be useful to help 3GPP in this=20
respect? Do you think a formal liaison would be useful to point it out?

3) 3GPP TS 36.300 is ambiguous whether ECN=20
marking is applied to all packets to "indicate=20
congestion", or whether it is applied with a=20
frequency or probability that depends on an AQM=20
(it doesn't mention AQM, altho obviously it=20
refers to RFC3168 that is based on AQM). Do you=20
know whether it was meant to imply use of AQM?

[Ruediger, thx for the supportive words too]


Bob

At 20:29 21/01/2014, Ingemar Johansson S wrote:
>Hi
>
>Please note that ECN over LTE radio access is=20
>already standardized in 3GPP TS 36.300 (see 11.6=20
>in http://www.3gpp.org/ftp/specs/archive/36_series/36.300/36300-c00.zip )
>
>/Ingemar
> > -----Original Message-----
> > From: Ruediger.Geib@telekom.de [mailto:Ruediger.Geib@telekom.de]
> > Sent: den 21 januari 2014 08:38
> > To: bob.briscoe@bt.com
> > Cc: tsvwg@ietf.org; aqm@ietf.org
> > Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to=
 L2
> > or tunnel protocols?
> >
> > Hi Bob,
> >
> > I support the issue being picked up by IETF. What can be done within the
> > bounds of IETF responsibility should be done. If ECN is seeing=
 deployment,
> > especially ECN support for IP over VLAN over IP/MPLS may be of interest.
> > Further, ECN over LTE radio Access may be relevant (but my expertise is=
 too
> > limited to judge details).
> >
> > Regards,
> >
> > Ruediger
> >
> > -----Urspr=FCngliche Nachricht-----
> > Von: tsvwg-bounces@ietf.org [mailto:tsvwg-bounces@ietf.org] Im Auftrag
> > von Bob Briscoe
> > Gesendet: Montag, 4. November 2013 23:04
> > An: tsvwg IETF list; AQM IETF list
> > Cc: draft-briscoe-tsvwg-ecn-encap-guidelines@tools.ietf.org
> > Betreff: [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or=
 tunnel
> > protocols?
> >
> > Folks,
> >
> > Pls respond if you support this being adopted as a work-group item in=
 the
> > IETF transport services w-g (tsvwg). The WG=20
> chairs need visibility of interest.
> > Even better, if you're willing to read / comment / review / implement
> >
> > Guidelines for Adding Congestion Notification to Protocols that=
 Encapsulate
> > IP <http://tools.ietf.org/html/draft-briscoe-tsvwg-ecn-encap-guidelines>
> >
> > Abstract
> >
> >     The purpose of this document is to guide the design of congestion
> >     notification in any lower layer or tunnelling protocol that
> >     encapsulates IP.  The aim is for explicit congestion signals to
> >     propagate consistently from lower layer protocols into IP.  Then the
> >     IP internetwork layer can act as a portability layer to carry
> >     congestion notification from non-IP-aware congested nodes up to the
> >     transport layer (L4).  Following these guidelines should assure
> >     interworking between new lower layer congestion notification
> >     mechanisms, whether specified by the IETF or other standards bodies.
> >
> >
> > [Cross-posting tsvwg & aqm, just in case]
> >
> >
> > Bob Briscoe,
> > also for co-authors Pat Thaler and John Kaippallimalil
> >
> >
> > __________________________________________________________
> > ______
> > Bob Briscoe,                                                  BT
> >
>
>_______________________________________________
>aqm mailing list
>aqm@ietf.org
>https://www.ietf.org/mailman/listinfo/aqm

________________________________________________________________
Bob Briscoe,                                                  BT=20


From preethi.cis@gmail.com  Tue Jan 21 16:42:41 2014
Return-Path: <preethi.cis@gmail.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5BAEF1A0233 for <aqm@ietfa.amsl.com>; Tue, 21 Jan 2014 16:42:41 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Cr7Q9wEDSDzP for <aqm@ietfa.amsl.com>; Tue, 21 Jan 2014 16:42:39 -0800 (PST)
Received: from mail-pa0-x234.google.com (mail-pa0-x234.google.com [IPv6:2607:f8b0:400e:c03::234]) by ietfa.amsl.com (Postfix) with ESMTP id C7B401A01BC for <aqm@ietf.org>; Tue, 21 Jan 2014 16:42:39 -0800 (PST)
Received: by mail-pa0-f52.google.com with SMTP id bj1so7630921pad.25 for <aqm@ietf.org>; Tue, 21 Jan 2014 16:42:39 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=user-agent:date:subject:from:to:message-id:thread-topic:in-reply-to :mime-version:content-type:content-transfer-encoding; bh=e96tNeqU5Qq/7wyd4k5dUkv4+yRYMbodjXRVeRCaqAg=; b=CcWXZoLnaUjKGBCo+qU6fnqbR6vKVHRfCCrt4LQdx2LVThNm9SjgS1FqRxTbN2bMZp 64h6RbUk0a/D62EpjxqlRWVcyGEdN50/6Q5t2++k1LAaFwzcZZbvXWDQhwOxZoUOaP2C Ze9Hef0MyEZY5AL9egxhIy07lmAf2rUcPImYXZFPR3/4FmwgGtBSBk0CDRxs3p+8ZCHH zyNjqEbhsvytmwdaw8vIOivz5JRoRGea/vyMrIFj9pcgZsqbZwBWuwr2///344zvHuWq Qm6zO0hKFK4u7pywq+3Ngi1mDFDRFfssmSR1Hk1F6AeGBGR0bwI6fSMyRi4hlym2PTd5 QyYg==
X-Received: by 10.68.75.9 with SMTP id y9mr28021717pbv.61.1390351359540; Tue, 21 Jan 2014 16:42:39 -0800 (PST)
Received: from [10.33.22.215] (128-107-239-234.cisco.com. [128.107.239.234]) by mx.google.com with ESMTPSA id yd4sm16266521pbc.13.2014.01.21.16.42.36 for <multiple recipients> (version=TLSv1 cipher=RC4-SHA bits=128/128); Tue, 21 Jan 2014 16:42:38 -0800 (PST)
User-Agent: Microsoft-MacOutlook/14.2.1.120420
Date: Tue, 21 Jan 2014 16:42:33 -0800
From: Preethi Natarajan <preethi.cis@gmail.com>
To: Nicolas KUHN <nicolas.kuhn@telecom-bretagne.eu>, <aqm@ietf.org>
Message-ID: <CF0433E6.4E751%prenatar@cisco.com>
Thread-Topic: [aqm]  Evalution guidelines - Editor team
In-Reply-To: <92D3FF75-D72D-4044-8796-E6B2A098DF5A@telecom-bretagne.eu>
Mime-version: 1.0
Content-type: text/plain; charset="US-ASCII"
Content-transfer-encoding: 7bit
Subject: Re: [aqm] Evalution guidelines - Editor team
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 Jan 2014 00:42:41 -0000

Hi Nicolas,

Welcome aboard the team.

I recently shared a preliminary draft with the rest of the editorial team.
The purpose of the draft was to gather thoughts in one place, iron-out the
categories for evaluation, and provide a starting point to kick off
discussions. I'll share that draft with you off-line.

Many thanks,
Preethi

On 1/17/14 12:59 AM, "Nicolas KUHN" <nicolas.kuhn@telecom-bretagne.eu>
wrote:

>Hi all, 
>
>I volunteer  to act as editor & contributor to the "AQM algorithm
>evaluation guidelines".  I just joined the RITE project [0].
>
>As explained in the email of Wes Eddy to the list [1], there was initial
>work presented in Vancouver [2] and we will write a draft out of it.
>
>>> Submit AQM algorithm evaluation guidelines to IESG for publication
>>> as Informational (Goal: July 2014)
>>>   - We need an editor team to step forward and begin work on this;
>>> there was initial work presented in Vancouver, but no draft available
>>>or
>>> adopted by the working group yet.
>
>
>A draft is coming really soon: we expect to send it by mid-February.
>
>Nicolas KUHN
>IMT
>
>[0] http://riteproject.eu
>[1] http://www.ietf.org/mail-archive/web/aqm/current/msg00461.html
>[2] http://www.ietf.org/proceedings/88/slides/slides-88-aqm-5.pdf
>_______________________________________________
>aqm mailing list
>aqm@ietf.org
>https://www.ietf.org/mailman/listinfo/aqm



From ingemar.s.johansson@ericsson.com  Tue Jan 21 22:36:30 2014
Return-Path: <ingemar.s.johansson@ericsson.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F14951A0237; Tue, 21 Jan 2014 22:36:29 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.851
X-Spam-Level: 
X-Spam-Status: No, score=-3.851 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HELO_EQ_SE=0.35, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Q0_mTOxCYwac; Tue, 21 Jan 2014 22:36:27 -0800 (PST)
Received: from mailgw1.ericsson.se (mailgw1.ericsson.se [193.180.251.45]) by ietfa.amsl.com (Postfix) with ESMTP id 5642F1A01AA; Tue, 21 Jan 2014 22:36:26 -0800 (PST)
X-AuditID: c1b4fb2d-b7f5d8e000002a7b-3d-52df66e910c3
Received: from ESESSHC024.ericsson.se (Unknown_Domain [153.88.253.124]) by mailgw1.ericsson.se (Symantec Mail Security) with SMTP id B3.91.10875.9E66FD25; Wed, 22 Jan 2014 07:36:25 +0100 (CET)
Received: from ESESSMB205.ericsson.se ([169.254.5.205]) by ESESSHC024.ericsson.se ([153.88.183.90]) with mapi id 14.02.0387.000; Wed, 22 Jan 2014 07:36:24 +0100
From: Ingemar Johansson S <ingemar.s.johansson@ericsson.com>
To: Bob Briscoe <bob.briscoe@bt.com>, "Ruediger.Geib@telekom.de" <Ruediger.Geib@telekom.de>
Thread-Topic: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to  L2 or tunnel protocols?
Thread-Index: AQHPFvsUyg4YEXg0W0GlvtH5Ta2mz5qQR/iQ
Date: Wed, 22 Jan 2014 06:36:24 +0000
Message-ID: <81564C0D7D4D2A4B9A86C8C7404A13DA31EC25B8@ESESSMB205.ericsson.se>
References: <201311042203.rA4M3lo0026458@bagheera.jungle.bt.co.uk> <CA7A7C64CC4ADB458B74477EA99DF6F502347353E1@HE111643.EMEA1.CDS.T-INTERNAL.COM> <81564C0D7D4D2A4B9A86C8C7404A13DA31EC250B@ESESSMB205.ericsson.se> <201401212249.s0LMneWw026346@bagheera.jungle.bt.co.uk>
In-Reply-To: <201401212249.s0LMneWw026346@bagheera.jungle.bt.co.uk>
Accept-Language: sv-SE, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [153.88.183.150]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFrrOLMWRmVeSWpSXmKPExsUyM+Jvje7LtPtBBg87+CzW7JO0mL7+C6PF h2kcFsfe3GVzYPFo+zKZyWPJkp9MHm0vFQKYo7hsUlJzMstSi/TtErgyNrXfZC7YpFlxbPVG tgbGV/JdjJwcEgImEkdXLGKHsMUkLtxbzwZiCwkcYpTYcUWmi5ELyF7CKLH233+wBJuAjcTK Q98ZQWwRgViJeX/uMXUxcnAwC7hITN3PDBIWFkiSeHu6mQWiJFniwuYdbCAlIgJGEvv+ZIGE WQRUJa5MamYCsXkFfCVuX5rFCrFqApPEjL0fWEESnALOEts6l4GtZRSQlbj//R7YTGYBcYlb T+YzQdwsILFkz3lmCFtU4uXjf6wQtpLEiu2XGCHq9SRuTJ3CBmFrSyxb+JoZYrGgxMmZT1gm MIrNQjJ2FpKWWUhaZiFpWcDIsoqRPTcxMye93HATIzByDm75rbuD8dQ5kUOM0hwsSuK8H946 BwkJpCeWpGanphakFsUXleakFh9iZOLglGpglD5UNGH7FOe0fccW+8Ue3t24tGlVWO7K5DIp g+qpHKK3mHYvN2708V0avqn50AF2FvcnyqkL7y9xytu/vTDoefaFE876DZeff9/SynpI50Rm dMHWeifButppq7pSNuRLBikH/T9/8Y1J6MeTssfVytwPe11RPOFuxvfQ/2rRzpRyYZ+2xyaN SizFGYmGWsxFxYkA2jLvRGoCAAA=
Cc: "tsvwg@ietf.org" <tsvwg@ietf.org>, "aqm@ietf.org" <aqm@ietf.org>
Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or tunnel protocols?
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 Jan 2014 06:36:30 -0000

Hi

Please find answers inline

/Ingemar

> -----Original Message-----
> From: Bob Briscoe [mailto:bob.briscoe@bt.com]
> Sent: den 21 januari 2014 23:50
> To: Ingemar Johansson S; Ruediger.Geib@telekom.de
> Cc: tsvwg@ietf.org; aqm@ietf.org
> Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to L=
2
> or tunnel protocols?
>=20
> Ingemar,
>=20
> 1) Thx for the pointer. We should add this as another example where a low=
er
> layer marks the IP header (the example we already have is the L3 switch i=
n
> the Ethernet world).
>=20
> 2) It doesn't talk about propagation of ECN markings during encap & decap=
,
> which is one of the things ecn-encap-guidelines aims to give guidelines o=
n.
> Do you think a doc like the draft we've done would be useful to help 3GPP=
 in
> this respect? Do you think a formal liaison would be useful to point it o=
ut?
L2 in this respect are the PDCP, RLC and MAC layers (to be honest I am not =
fully clear what constitutes L2). The problem is that to given encap/decap =
a meaning in this case it becomes necessary to encode at least an ECN-CE bi=
t into any of the PDCP or RLC or MAC headers, as PDCP is ciphered it probab=
ly needs to be in the RLC or MAC headers. Currently there are no standards =
efforts to add an ECN-CE bit to any of these headers.

>=20
> 3) 3GPP TS 36.300 is ambiguous whether ECN marking is applied to all pack=
ets
> to "indicate congestion", or whether it is applied with a frequency or
> probability that depends on an AQM (it doesn't mention AQM, altho
> obviously it refers to RFC3168 that is based on AQM). Do you know whether
> it was meant to imply use of AQM?
Yes, it is very ambiguous, I advocate for a frequency or probability as it =
makes most sense, but this is not specified at all. The ECN marking can imp=
ly the use of AQM but that is again not specified. In short, much of the ac=
tual implementation is vendor specific.

>=20
> [Ruediger, thx for the supportive words too]
>=20
>=20
> Bob
>=20
> At 20:29 21/01/2014, Ingemar Johansson S wrote:
> >Hi
> >
> >Please note that ECN over LTE radio access is already standardized in
> >3GPP TS 36.300 (see 11.6 in
> >http://www.3gpp.org/ftp/specs/archive/36_series/36.300/36300-c00.zip )
> >
> >/Ingemar
> > > -----Original Message-----
> > > From: Ruediger.Geib@telekom.de [mailto:Ruediger.Geib@telekom.de]
> > > Sent: den 21 januari 2014 08:38
> > > To: bob.briscoe@bt.com
> > > Cc: tsvwg@ietf.org; aqm@ietf.org
> > > Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN
> > > to L2 or tunnel protocols?
> > >
> > > Hi Bob,
> > >
> > > I support the issue being picked up by IETF. What can be done within
> > > the bounds of IETF responsibility should be done. If ECN is seeing
> > > deployment, especially ECN support for IP over VLAN over IP/MPLS may
> be of interest.
> > > Further, ECN over LTE radio Access may be relevant (but my expertise
> > > is too limited to judge details).
> > >
> > > Regards,
> > >
> > > Ruediger
> > >
> > > -----Urspr=FCngliche Nachricht-----
> > > Von: tsvwg-bounces@ietf.org [mailto:tsvwg-bounces@ietf.org] Im
> > > Auftrag von Bob Briscoe
> > > Gesendet: Montag, 4. November 2013 23:04
> > > An: tsvwg IETF list; AQM IETF list
> > > Cc: draft-briscoe-tsvwg-ecn-encap-guidelines@tools.ietf.org
> > > Betreff: [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or
> > > tunnel protocols?
> > >
> > > Folks,
> > >
> > > Pls respond if you support this being adopted as a work-group item
> > > in the IETF transport services w-g (tsvwg). The WG
> > chairs need visibility of interest.
> > > Even better, if you're willing to read / comment / review /
> > > implement
> > >
> > > Guidelines for Adding Congestion Notification to Protocols that
> > > Encapsulate IP
> > > <http://tools.ietf.org/html/draft-briscoe-tsvwg-ecn-encap-guidelines
> > > >
> > >
> > > Abstract
> > >
> > >     The purpose of this document is to guide the design of congestion
> > >     notification in any lower layer or tunnelling protocol that
> > >     encapsulates IP.  The aim is for explicit congestion signals to
> > >     propagate consistently from lower layer protocols into IP.  Then =
the
> > >     IP internetwork layer can act as a portability layer to carry
> > >     congestion notification from non-IP-aware congested nodes up to t=
he
> > >     transport layer (L4).  Following these guidelines should assure
> > >     interworking between new lower layer congestion notification
> > >     mechanisms, whether specified by the IETF or other standards bodi=
es.
> > >
> > >
> > > [Cross-posting tsvwg & aqm, just in case]
> > >
> > >
> > > Bob Briscoe,
> > > also for co-authors Pat Thaler and John Kaippallimalil
> > >
> > >
> > >
> __________________________________________________________
> > > ______
> > > Bob Briscoe,                                                  BT
> > >
> >
> >_______________________________________________
> >aqm mailing list
> >aqm@ietf.org
> >https://www.ietf.org/mailman/listinfo/aqm
>=20
> __________________________________________________________
> ______
> Bob Briscoe,                                                  BT


From fred@cisco.com  Wed Jan 22 22:10:29 2014
Return-Path: <fred@cisco.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7F1881A0200 for <aqm@ietfa.amsl.com>; Wed, 22 Jan 2014 22:10:29 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -115.035
X-Spam-Level: 
X-Spam-Status: No, score=-115.035 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.535, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5, USER_IN_WHITELIST=-100] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id frBh3XAx7J8v for <aqm@ietfa.amsl.com>; Wed, 22 Jan 2014 22:10:26 -0800 (PST)
Received: from rcdn-iport-5.cisco.com (rcdn-iport-5.cisco.com [173.37.86.76]) by ietfa.amsl.com (Postfix) with ESMTP id A67791A0214 for <aqm@ietf.org>; Wed, 22 Jan 2014 22:10:25 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=13091; q=dns/txt; s=iport; t=1390457425; x=1391667025; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=ralj3x/Uz+ovBuJRZ4V8Wdn/KVlIenLa2P26xn/Eqdw=; b=G5e1XnoRUu3S/+E+D9bGKdSPAV7g5ir5VpB7r8SUEbDm2CS9gesVxHZL KXlXb4wRsbIhs3cW6Bk7WxTGuuR1Qj7JCcTnkD2FocDGyPRkkmCDiztwh UWvvM4ZbLJ3D7qLXkytjY6yymLNtbDhniF7jqfokvi+lRDoJ2QS88XkK/ s=;
X-Files: signature.asc : 195
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: AgkFAMyx4FKtJXHA/2dsb2JhbABYA4MMOFapbpIEgRAWdIIlAQEBAwEBAQFrCwULAgEIRicLJQIEDgUOh28IDcRiEwSOGVMQBxGDE4EUBJA7gTGGNpIYgW+BPoIq
X-IronPort-AV: E=Sophos;i="4.95,704,1384300800";  d="asc'?scan'208,217";a="299129615"
Received: from rcdn-core2-5.cisco.com ([173.37.113.192]) by rcdn-iport-5.cisco.com with ESMTP; 23 Jan 2014 06:10:24 +0000
Received: from xhc-aln-x10.cisco.com (xhc-aln-x10.cisco.com [173.36.12.84]) by rcdn-core2-5.cisco.com (8.14.5/8.14.5) with ESMTP id s0N6AOwB005655 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 23 Jan 2014 06:10:24 GMT
Received: from xmb-rcd-x09.cisco.com ([169.254.9.230]) by xhc-aln-x10.cisco.com ([173.36.12.84]) with mapi id 14.03.0123.003; Thu, 23 Jan 2014 00:10:24 -0600
From: "Fred Baker (fred)" <fred@cisco.com>
To: "<davecb@spamcop.net>" <davecb@spamcop.net>
Thread-Topic: [aqm] Prefatory comments re draft-aqm-reccommendation and -evaluation, and a question
Thread-Index: AQHPGAHNbKCMVOEBQkeCeZV6ncqOcQ==
Date: Thu, 23 Jan 2014 06:10:23 +0000
Message-ID: <4FD254F4-D1E3-4CD1-AE35-C29C0DD37B4B@cisco.com>
References: <52D9C1F3.3070803@rogers.com>
In-Reply-To: <52D9C1F3.3070803@rogers.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
x-originating-ip: [10.19.64.117]
Content-Type: multipart/signed; boundary="Apple-Mail=_FCCBABEA-37A4-4D63-A262-D4BF72073E9E"; protocol="application/pgp-signature"; micalg=pgp-sha1
MIME-Version: 1.0
Cc: "aqm@ietf.org" <aqm@ietf.org>, Dave Taht <dave.taht@gmail.com>
Subject: Re: [aqm] Prefatory comments re draft-aqm-reccommendation and -evaluation, and a question
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 Jan 2014 06:10:30 -0000

--Apple-Mail=_FCCBABEA-37A4-4D63-A262-D4BF72073E9E
Content-Type: multipart/alternative;
	boundary="Apple-Mail=_3F083B51-2162-4EAB-B524-00F6DE57FD12"


--Apple-Mail=_3F083B51-2162-4EAB-B524-00F6DE57FD12
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii

No, you're not blowing smoke. I'm not sure I would compare the behavior =
to PMTUD, as in that the endpoint is given a magic number and manages to =
it, where in this case, it is given the results of its behavior, and it =
manages to improve that.=20

But this is what I have rambled on about in threads relating to the size =
of a buffer. Folks would really like to have a magic way to calculate =
the buffer size (an amount of memory) they need to install in a router =
or switch, and it isn't that easy, because it has a lot to do with where =
in the network a system is located and how it is used by the =
applications that use it. But AQM, in the end, isn't about buffer size. =
It is about buffer occupancy. In the ideal case, if there are N sessions =
active on the bottleneck link in a path, we would like each to obtain =
1/N of the bottleneck's capacity, which is to say that it should be able =
to maximize its throughput, while keeping an average of zero packets =
standing in the queue (minimizing both latency and variation in =
latency). If you know your math, you know that the ideal goal isn't =
actually achievable. But that doesn't stop us from trying to =
asymptotically approach it.

On Jan 17, 2014, at 3:51 PM, David Collier-Brown <davec-b@rogers.com> =
wrote:

> I've been reading through the internet-drafts, and one paragraph =
struck me as very illuminating.  This is therefor a sanity-check before =
I go full-hog down a particular path...
>=20
> The comment is from Baker and Fairhurst, =
https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/ and the =
paragraph is [emphases added]
> The point of buffering in the network is to absorb data bursts and to =
transmit them during the (hopefully) ensuing bursts of silence. This is =
essential to permit the transmission of bursty data. Normally small =
queues are preferred in network devices, with sufficient queue capacity =
to absorb the bursts. The counter-intuitive result is that maintaining =
normally-small queues can result in higher throughput as well as lower =
end-to- end delay. In summary, queue limits should not reflect the =
steady state queues we want to be maintained in the network; instead, =
they should reflect the size of bursts that a network device needs to =
absorb.=20
> All of a sudden we're talking about the kinds of queues I know a =
little about (:-))
> ---
>=20
> I'm going to suggest that these are queues and associated physical =
buffers that do two things:
> hold packets that arrive at a bottleneck for a long as it takes to =
send them out a slower link that they came in on, and
> hold bursts of packets that arrive adjacent to each other until they =
can be sent out in a normal spacing, with some small amount of time =
between them
>=20
> In an illustration of Dave Taht's, the first looks something like this
>=20
> -------------------------+
>     |X|            |X|      +-------------------
>     |X|            |X|         |XXX|    |XXX|
>     |X|            |X|      +-------------------
> -------------------------+
>=20
> At the choke-point there is a buffer at least big enough to give the =
packet a chance to wheel from line into column (:-)) and start down     =
the smaller pipe.
>=20
> The speed at which the acks come back,  the frequency of drops, and =
any explicit congestion notifications slows the sender until they don't =
overload the skinnier pipe, thus spacing the packets in the fatter pipe =
out.
>=20
> Various causes [Leland] can slow or speed the packets in the fat pipe, =
making it possible for several to arrive adjacent to each other, =
followed by a gap.  The second purpose of a buffer is to hold  these =
bursts while things space themselves back out.
>=20
> They need to be big enough at minimum to do the speed matching,  and =
at maximum,  big enough to spread a burst back into a normal =
progression, always assuming that acks, drops and explicit congestion =
notifications are slowing the sender to the speed of the slowest part of =
the network.
> ---
>=20
> If I'm right about this, we can draw some helpful conclusions
> buffer sizes can be set based on measurements:
> speed differences, which are pretty static, plus
> observed burstyness
> drops and ECN can be done to match the slowest speed in the path
> The latter suddenly sounds a bit like path MTU discovery, except it's =
a bit more dynamic, and varies with both path and what's happening in =
various parts of it.
> To me, as a capacity/performance nerd, this sounds a lot more familiar =
and manageable. My question to you, before I start madly scribbling on =
the internet drafts is:
>     Am I blowing smoke?
> --dave
> --=20
> David Collier-Brown,         | Always do right. This will gratify
> System Programmer and Author | some people and astonish the rest
> davecb@spamcop.net           |                      -- Mark Twain
> (416) 223-8968
> _______________________________________________
> aqm mailing list
> aqm@ietf.org
> https://www.ietf.org/mailman/listinfo/aqm



--Apple-Mail=_3F083B51-2162-4EAB-B524-00F6DE57FD12
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=us-ascii

<html><head><meta http-equiv=3D"Content-Type" content=3D"text/html =
charset=3Dus-ascii"></head><body style=3D"word-wrap: break-word; =
-webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">No, =
you're not blowing smoke. I'm not sure I would compare the behavior to =
PMTUD, as in that the endpoint is given a magic number and manages to =
it, where in this case, it is given the results of its behavior, and it =
manages to improve that.&nbsp;<div><br></div><div>But this is what I =
have rambled on about in threads relating to the size of a buffer. Folks =
would really like to have a magic way to calculate the buffer size (an =
amount of memory) they need to install in a router or switch, and it =
isn't that easy, because it has a lot to do with where in the network a =
system is located and how it is used by the applications that use it. =
But AQM, in the end, isn't about buffer size. It is about buffer =
occupancy.&nbsp;In the ideal case, if there are N sessions active on the =
bottleneck link in a path, we would like each to obtain 1/N of the =
bottleneck's capacity, which is to say that it should be able to =
maximize its throughput, while keeping an average of zero packets =
standing in the queue (minimizing both latency and variation in =
latency). If you know your math, you know that the ideal goal isn't =
actually achievable. But that doesn't stop us from trying to =
asymptotically approach it.</div><div><br><div><div>On Jan 17, 2014, at =
3:51 PM, David Collier-Brown &lt;<a =
href=3D"mailto:davec-b@rogers.com">davec-b@rogers.com</a>&gt; =
wrote:</div><br class=3D"Apple-interchange-newline"><blockquote =
type=3D"cite">
<meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3Diso-8859-1">
 =20
  <div bgcolor=3D"#FFFFFF" text=3D"#000000">
    I've been reading through the internet-drafts, and one paragraph
    struck me as very illuminating.&nbsp; This is therefor a =
sanity-check
    before I go full-hog down a particular path...<br>
    <br>
    The comment is from Baker and Fairhurst,
    <a class=3D"moz-txt-link-freetext" =
href=3D"https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/">h=
ttps://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/</a> and
    the paragraph is [emphases added]<br>
    <blockquote> The point of buffering in the network is to absorb data
      bursts and to transmit them during the (hopefully) ensuing bursts
      of silence. This is essential to permit the transmission of bursty
      data. Normally small queues are preferred in network devices, with
      sufficient queue capacity to absorb the bursts. The
      counter-intuitive result is that maintaining normally-small queues
      can result in higher throughput as well as lower end-to- end
      delay. In summary, queue limits should not reflect the steady
      state queues we want to be maintained in the network; i<i>nstead,
        they should reflect the size of bursts that a network device
        needs to absorb.
      </i><br>
    </blockquote>
    All of a sudden we're talking about the kinds of queues I know a
    little about (:-))<br>
    ---<br>
    <br>
    I'm going to suggest that these are queues and associated physical
    buffers that do two things:<br>
    <ol>
      <li>hold packets that arrive at a bottleneck for a long as it
        takes to send them out a slower link that they came in on, and =
</li>
      <li>hold bursts of packets that arrive adjacent to each other
        until they can be sent out in a normal spacing, with some small
        amount of time between them</li>
    </ol>
    <br>
    In an illustration of Dave Taht's, the first looks something like
    this<br>
    <br>
    -------------------------+<br>
    &nbsp;&nbsp;&nbsp; =
|X|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
|X|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; +-------------------<br>
    &nbsp;&nbsp;&nbsp; =
|X|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
|X|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
|XXX|&nbsp;&nbsp;&nbsp; |XXX|<br>
    &nbsp;&nbsp;&nbsp; =
|X|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
|X|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; +-------------------<br>
    -------------------------+<br>
    <br>
    At the choke-point there is a buffer at least big enough to give the
    packet a chance to wheel from line into column (:-)) and start down
    the smaller pipe.<br>
    <br>
    The speed at which the acks come back,&nbsp; the frequency of drops, =
and
    any explicit congestion notifications slows the sender until they
    don't overload the skinnier pipe, thus spacing the packets in the
    fatter pipe out.<br>
    <br>
    Various causes [Leland] can slow or speed the packets in the fat
    pipe, making it possible for several to arrive adjacent to each
    other, followed by a gap.&nbsp; The second purpose of a buffer is to
    hold&nbsp; these bursts while things space themselves back out.<br>
    <br>
    They need to be big enough at minimum to do the speed =
matching,&nbsp; and
    at maximum,&nbsp; big enough to spread a burst back into a normal
    progression, always assuming that acks, drops and explicit
    congestion notifications are slowing the sender to the speed of the
    slowest part of the network.<br>
    ---<br>
    <br>
    If I'm right about this, we can draw some helpful conclusions<br>
    <ul>
      <li>buffer sizes can be set based on measurements:</li>
      <ul>
        <li>speed differences, which are pretty static, plus</li>
        <li>observed burstyness<br>
        </li>
      </ul>
      <li>drops and ECN can be done to match the slowest speed in the
        path<br>
      </li>
    </ul><p>The latter suddenly sounds a bit like path MTU discovery, =
except
      it's a bit more dynamic, and varies with both path and what's
      happening in various parts of it.<br>
    </p><p>To me, as a capacity/performance nerd, this sounds a lot more
      familiar and manageable. My question to you, before I start madly
      scribbling on the internet drafts is:<br>
    </p><p>&nbsp;&nbsp;&nbsp; <i>Am I blowing smoke?</i><br>
    </p><p>--dave<br>
    </p>
    <pre class=3D"moz-signature" cols=3D"72">--=20
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
<a class=3D"moz-txt-link-abbreviated" =
href=3D"mailto:davecb@spamcop.net">davecb@spamcop.net</a>           |    =
                  -- Mark Twain
(416) 223-8968
</pre>
  </div>

_______________________________________________<br>aqm mailing =
list<br><a =
href=3D"mailto:aqm@ietf.org">aqm@ietf.org</a><br>https://www.ietf.org/mail=
man/listinfo/aqm<br></blockquote></div><br>
<br></div></body></html>=

--Apple-Mail=_3F083B51-2162-4EAB-B524-00F6DE57FD12--

--Apple-Mail=_FCCBABEA-37A4-4D63-A262-D4BF72073E9E
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment; filename="signature.asc"
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Message signed with OpenPGP using GPGMail

-----BEGIN PGP SIGNATURE-----
Comment: GPGTools - http://gpgtools.org

iD8DBQFS4LJMbjEdbHIsm0MRAsbzAKCmd6+2TrY5CxVF3LjCteYyOLMkPgCgqzYb
VAUBEol/hRqtStSImKkJ0o8=
=YvK7
-----END PGP SIGNATURE-----

--Apple-Mail=_FCCBABEA-37A4-4D63-A262-D4BF72073E9E--

From dave.taht@gmail.com  Wed Jan 22 22:15:06 2014
Return-Path: <dave.taht@gmail.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C39681A024F for <aqm@ietfa.amsl.com>; Wed, 22 Jan 2014 22:15:06 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tXM4nLImcj2H for <aqm@ietfa.amsl.com>; Wed, 22 Jan 2014 22:15:04 -0800 (PST)
Received: from mail-ie0-x22d.google.com (mail-ie0-x22d.google.com [IPv6:2607:f8b0:4001:c03::22d]) by ietfa.amsl.com (Postfix) with ESMTP id D5B231A0249 for <aqm@ietf.org>; Wed, 22 Jan 2014 22:15:03 -0800 (PST)
Received: by mail-ie0-f173.google.com with SMTP id e14so672616iej.4 for <aqm@ietf.org>; Wed, 22 Jan 2014 22:15:03 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=StZTAvldvwYzVg4k4weg9Rhhh7PXvl06sEWt2zCNKo4=; b=i8bxQZjZ95KMtWVdgYreQgErJMuOc9FRv6JGxEl3MCIxaJ10TKg+TGxYyUIK4B8otc XBeHeiJRPD5JQXRneWEcS7jTopMC9QF2EJqs6KnkPsrNsrXt2cjzBloMPsqlGtBlCbaa ziUryC84wP7XB8+HRXWGqZfqR+vEBCHCRCqbO+ifPlMXHo6Z4EnHAdPMznmOftR/k8ni dYySG4f4tPaR1vjKMX56b6DE4xcqQfor3f7bI8TcTMouZXrCcOl4LMmFa2YT+juvn2Sb aAWtPWApHTKZsostPFr/rDwi+1GNC0jt1r0FNsIuljTHyQ34gkVi+fe0Zn2DkXsZCycB XG5w==
MIME-Version: 1.0
X-Received: by 10.50.36.67 with SMTP id o3mr6842071igj.47.1390457703054; Wed, 22 Jan 2014 22:15:03 -0800 (PST)
Received: by 10.64.145.67 with HTTP; Wed, 22 Jan 2014 22:15:02 -0800 (PST)
In-Reply-To: <4FD254F4-D1E3-4CD1-AE35-C29C0DD37B4B@cisco.com>
References: <52D9C1F3.3070803@rogers.com> <4FD254F4-D1E3-4CD1-AE35-C29C0DD37B4B@cisco.com>
Date: Thu, 23 Jan 2014 01:15:02 -0500
Message-ID: <CAA93jw4Yw=SHQYKMs5_LaRHYa18RYeKgo2gS5nET3XvhwR8Pjg@mail.gmail.com>
From: Dave Taht <dave.taht@gmail.com>
To: "Fred Baker (fred)" <fred@cisco.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: "aqm@ietf.org" <aqm@ietf.org>, "<davecb@spamcop.net>" <davecb@spamcop.net>
Subject: Re: [aqm] Prefatory comments re draft-aqm-reccommendation and -evaluation, and a question
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 Jan 2014 06:15:07 -0000

On Thu, Jan 23, 2014 at 1:10 AM, Fred Baker (fred) <fred@cisco.com> wrote:
> No, you're not blowing smoke. I'm not sure I would compare the behavior t=
o
> PMTUD, as in that the endpoint is given a magic number and manages to it,
> where in this case, it is given the results of its behavior, and it manag=
es
> to improve that.
>
> But this is what I have rambled on about in threads relating to the size =
of
> a buffer. Folks would really like to have a magic way to calculate the
> buffer size (an amount of memory) they need to install in a router or
> switch, and it isn't that easy, because it has a lot to do with where in =
the
> network a system is located and how it is used by the applications that u=
se
> it. But AQM, in the end, isn't about buffer size. It is about buffer
> occupancy. In the ideal case, if there are N sessions active on the
> bottleneck link in a path, we would like each to obtain 1/N of the
> bottleneck's capacity, which is to say that it should be able to maximize
> its throughput, while keeping an average of zero packets standing in the
> queue (minimizing both latency and variation in latency). If you know you=
r
> math, you know that the ideal goal isn't actually achievable. But that
> doesn't stop us from trying to asymptotically approach it.

I prefer to think of the goal as to keep a minimum of 1 packet in the queue=
,
not as an average of 0.

>
> On Jan 17, 2014, at 3:51 PM, David Collier-Brown <davec-b@rogers.com> wro=
te:
>
> I've been reading through the internet-drafts, and one paragraph struck m=
e
> as very illuminating.  This is therefor a sanity-check before I go full-h=
og
> down a particular path...
>
> The comment is from Baker and Fairhurst,
> https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/ and the
> paragraph is [emphases added]
>
> The point of buffering in the network is to absorb data bursts and to
> transmit them during the (hopefully) ensuing bursts of silence. This is
> essential to permit the transmission of bursty data. Normally small queue=
s
> are preferred in network devices, with sufficient queue capacity to absor=
b
> the bursts. The counter-intuitive result is that maintaining normally-sma=
ll
> queues can result in higher throughput as well as lower end-to- end delay=
.
> In summary, queue limits should not reflect the steady state queues we wa=
nt
> to be maintained in the network; instead, they should reflect the size of
> bursts that a network device needs to absorb.
>
> All of a sudden we're talking about the kinds of queues I know a little
> about (:-))
> ---
>
> I'm going to suggest that these are queues and associated physical buffer=
s
> that do two things:
>
> hold packets that arrive at a bottleneck for a long as it takes to send t=
hem
> out a slower link that they came in on, and
> hold bursts of packets that arrive adjacent to each other until they can =
be
> sent out in a normal spacing, with some small amount of time between them
>
>
> In an illustration of Dave Taht's, the first looks something like this
>
> -------------------------+
>     |X|            |X|      +-------------------
>     |X|            |X|         |XXX|    |XXX|
>     |X|            |X|      +-------------------
> -------------------------+
>
> At the choke-point there is a buffer at least big enough to give the pack=
et
> a chance to wheel from line into column (:-)) and start down the smaller
> pipe.
>
> The speed at which the acks come back,  the frequency of drops, and any
> explicit congestion notifications slows the sender until they don't overl=
oad
> the skinnier pipe, thus spacing the packets in the fatter pipe out.
>
> Various causes [Leland] can slow or speed the packets in the fat pipe,
> making it possible for several to arrive adjacent to each other, followed=
 by
> a gap.  The second purpose of a buffer is to hold  these bursts while thi=
ngs
> space themselves back out.
>
> They need to be big enough at minimum to do the speed matching,  and at
> maximum,  big enough to spread a burst back into a normal progression,
> always assuming that acks, drops and explicit congestion notifications ar=
e
> slowing the sender to the speed of the slowest part of the network.
> ---
>
> If I'm right about this, we can draw some helpful conclusions
>
> buffer sizes can be set based on measurements:
>
> speed differences, which are pretty static, plus
> observed burstyness
>
> drops and ECN can be done to match the slowest speed in the path
>
> The latter suddenly sounds a bit like path MTU discovery, except it's a b=
it
> more dynamic, and varies with both path and what's happening in various
> parts of it.
>
> To me, as a capacity/performance nerd, this sounds a lot more familiar an=
d
> manageable. My question to you, before I start madly scribbling on the
> internet drafts is:
>
>     Am I blowing smoke?
>
> --dave
>
> --
> David Collier-Brown,         | Always do right. This will gratify
> System Programmer and Author | some people and astonish the rest
> davecb@spamcop.net           |                      -- Mark Twain
> (416) 223-8968
>
> _______________________________________________
>
> aqm mailing list
> aqm@ietf.org
> https://www.ietf.org/mailman/listinfo/aqm
>
>
>



--=20
Dave T=E4ht

Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.=
html

From dave.taht@gmail.com  Wed Jan 22 22:18:28 2014
Return-Path: <dave.taht@gmail.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 63FE51A016D for <aqm@ietfa.amsl.com>; Wed, 22 Jan 2014 22:18:28 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WxrK4aeh13Z6 for <aqm@ietfa.amsl.com>; Wed, 22 Jan 2014 22:18:25 -0800 (PST)
Received: from mail-ie0-x233.google.com (mail-ie0-x233.google.com [IPv6:2607:f8b0:4001:c03::233]) by ietfa.amsl.com (Postfix) with ESMTP id 626071A0238 for <aqm@ietf.org>; Wed, 22 Jan 2014 22:18:25 -0800 (PST)
Received: by mail-ie0-f179.google.com with SMTP id ar20so677172iec.38 for <aqm@ietf.org>; Wed, 22 Jan 2014 22:18:24 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=DexBAMoBHqNlCKAqbGvlTGqePtXhlSBdo+0/ISMcuas=; b=xsgV08Gd39+1wy48yOE+Stpi1/lbhd8VXPuxZGYXR3tmQiVfXmBbe5Kjl3bLv1xUOK 3S5B9u3Rev6Ccn/waLekz92LZJfgwzfGHI6RAYOI55/1kHeRB66jqBG3UXRtScwZTuQ2 881DbQdlWAFAABa9QN8bB8qFx0apOGQpZmIJFCAHCeyH6jTlQq61/TswgCNZfxkYVpy5 LrdsQ/WQTgJHGJu3gZ/Mp9/Qzy8+/wIX9Nl0y4qYlzxtobzmRvQdq8ARB77Iz2QEIbvZ OlR6IcA7hUptIw0vOSUzz1WMJ/yP6fBWm8XiOvg9ttmCWCLULnKRlPJ7W7MVlLp2Ly8K ouVg==
MIME-Version: 1.0
X-Received: by 10.50.36.67 with SMTP id o3mr6855692igj.47.1390457904556; Wed, 22 Jan 2014 22:18:24 -0800 (PST)
Received: by 10.64.145.67 with HTTP; Wed, 22 Jan 2014 22:18:24 -0800 (PST)
In-Reply-To: <CAA93jw4Yw=SHQYKMs5_LaRHYa18RYeKgo2gS5nET3XvhwR8Pjg@mail.gmail.com>
References: <52D9C1F3.3070803@rogers.com> <4FD254F4-D1E3-4CD1-AE35-C29C0DD37B4B@cisco.com> <CAA93jw4Yw=SHQYKMs5_LaRHYa18RYeKgo2gS5nET3XvhwR8Pjg@mail.gmail.com>
Date: Thu, 23 Jan 2014 01:18:24 -0500
Message-ID: <CAA93jw4Goh3Y4OrwTJDUcBGN0bj6YEAm++MYuAaqXiS4Er6q+g@mail.gmail.com>
From: Dave Taht <dave.taht@gmail.com>
To: "Fred Baker (fred)" <fred@cisco.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: "aqm@ietf.org" <aqm@ietf.org>, "<davecb@spamcop.net>" <davecb@spamcop.net>
Subject: Re: [aqm] Prefatory comments re draft-aqm-reccommendation and -evaluation, and a question
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 Jan 2014 06:18:28 -0000

On Thu, Jan 23, 2014 at 1:15 AM, Dave Taht <dave.taht@gmail.com> wrote:
> On Thu, Jan 23, 2014 at 1:10 AM, Fred Baker (fred) <fred@cisco.com> wrote=
:
>> No, you're not blowing smoke. I'm not sure I would compare the behavior =
to
>> PMTUD, as in that the endpoint is given a magic number and manages to it=
,
>> where in this case, it is given the results of its behavior, and it mana=
ges
>> to improve that.
>>
>> But this is what I have rambled on about in threads relating to the size=
 of
>> a buffer. Folks would really like to have a magic way to calculate the
>> buffer size (an amount of memory) they need to install in a router or
>> switch, and it isn't that easy, because it has a lot to do with where in=
 the
>> network a system is located and how it is used by the applications that =
use
>> it. But AQM, in the end, isn't about buffer size. It is about buffer
>> occupancy. In the ideal case, if there are N sessions active on the
>> bottleneck link in a path, we would like each to obtain 1/N of the
>> bottleneck's capacity, which is to say that it should be able to maximiz=
e
>> its throughput, while keeping an average of zero packets standing in the
>> queue (minimizing both latency and variation in latency). If you know yo=
ur
>> math, you know that the ideal goal isn't actually achievable. But that
>> doesn't stop us from trying to asymptotically approach it.
>
> I prefer to think of the goal as to keep a minimum of 1 packet in the que=
ue,
> not as an average of 0.

And that's not strictly true either. In the case of wifi, and other bundlin=
g
technologies like those used in cable, you want to keep a minimum of a "goo=
d"
aggregate of packets in the queue for that technology.

>
>>
>> On Jan 17, 2014, at 3:51 PM, David Collier-Brown <davec-b@rogers.com> wr=
ote:
>>
>> I've been reading through the internet-drafts, and one paragraph struck =
me
>> as very illuminating.  This is therefor a sanity-check before I go full-=
hog
>> down a particular path...
>>
>> The comment is from Baker and Fairhurst,
>> https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/ and the
>> paragraph is [emphases added]
>>
>> The point of buffering in the network is to absorb data bursts and to
>> transmit them during the (hopefully) ensuing bursts of silence. This is
>> essential to permit the transmission of bursty data. Normally small queu=
es
>> are preferred in network devices, with sufficient queue capacity to abso=
rb
>> the bursts. The counter-intuitive result is that maintaining normally-sm=
all
>> queues can result in higher throughput as well as lower end-to- end dela=
y.
>> In summary, queue limits should not reflect the steady state queues we w=
ant
>> to be maintained in the network; instead, they should reflect the size o=
f
>> bursts that a network device needs to absorb.
>>
>> All of a sudden we're talking about the kinds of queues I know a little
>> about (:-))
>> ---
>>
>> I'm going to suggest that these are queues and associated physical buffe=
rs
>> that do two things:
>>
>> hold packets that arrive at a bottleneck for a long as it takes to send =
them
>> out a slower link that they came in on, and
>> hold bursts of packets that arrive adjacent to each other until they can=
 be
>> sent out in a normal spacing, with some small amount of time between the=
m
>>
>>
>> In an illustration of Dave Taht's, the first looks something like this
>>
>> -------------------------+
>>     |X|            |X|      +-------------------
>>     |X|            |X|         |XXX|    |XXX|
>>     |X|            |X|      +-------------------
>> -------------------------+
>>
>> At the choke-point there is a buffer at least big enough to give the pac=
ket
>> a chance to wheel from line into column (:-)) and start down the smaller
>> pipe.
>>
>> The speed at which the acks come back,  the frequency of drops, and any
>> explicit congestion notifications slows the sender until they don't over=
load
>> the skinnier pipe, thus spacing the packets in the fatter pipe out.
>>
>> Various causes [Leland] can slow or speed the packets in the fat pipe,
>> making it possible for several to arrive adjacent to each other, followe=
d by
>> a gap.  The second purpose of a buffer is to hold  these bursts while th=
ings
>> space themselves back out.
>>
>> They need to be big enough at minimum to do the speed matching,  and at
>> maximum,  big enough to spread a burst back into a normal progression,
>> always assuming that acks, drops and explicit congestion notifications a=
re
>> slowing the sender to the speed of the slowest part of the network.
>> ---
>>
>> If I'm right about this, we can draw some helpful conclusions
>>
>> buffer sizes can be set based on measurements:
>>
>> speed differences, which are pretty static, plus
>> observed burstyness
>>
>> drops and ECN can be done to match the slowest speed in the path
>>
>> The latter suddenly sounds a bit like path MTU discovery, except it's a =
bit
>> more dynamic, and varies with both path and what's happening in various
>> parts of it.
>>
>> To me, as a capacity/performance nerd, this sounds a lot more familiar a=
nd
>> manageable. My question to you, before I start madly scribbling on the
>> internet drafts is:
>>
>>     Am I blowing smoke?
>>
>> --dave
>>
>> --
>> David Collier-Brown,         | Always do right. This will gratify
>> System Programmer and Author | some people and astonish the rest
>> davecb@spamcop.net           |                      -- Mark Twain
>> (416) 223-8968
>>
>> _______________________________________________
>>
>> aqm mailing list
>> aqm@ietf.org
>> https://www.ietf.org/mailman/listinfo/aqm
>>
>>
>>
>
>
>
> --
> Dave T=E4ht
>
> Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscrib=
e.html



--=20
Dave T=E4ht

Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.=
html

From davec-b@rogers.com  Thu Jan 23 05:59:40 2014
Return-Path: <davec-b@rogers.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8CC211A00F2 for <aqm@ietfa.amsl.com>; Thu, 23 Jan 2014 05:59:40 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.399
X-Spam-Level: 
X-Spam-Status: No, score=-1.399 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, J_CHICKENPOX_64=0.6, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id spkElnIFtrGh for <aqm@ietfa.amsl.com>; Thu, 23 Jan 2014 05:59:36 -0800 (PST)
Received: from nm6-vm8.access.bullet.mail.gq1.yahoo.com (nm6-vm8.access.bullet.mail.gq1.yahoo.com [216.39.63.214]) by ietfa.amsl.com (Postfix) with ESMTP id D151C1A0434 for <aqm@ietf.org>; Thu, 23 Jan 2014 05:59:36 -0800 (PST)
Received: from [216.39.60.174] by nm6.access.bullet.mail.gq1.yahoo.com with NNFMP; 23 Jan 2014 13:59:36 -0000
Received: from [67.195.23.145] by tm10.access.bullet.mail.gq1.yahoo.com with NNFMP; 23 Jan 2014 13:59:35 -0000
Received: from [127.0.0.1] by smtp117.sbc.mail.gq1.yahoo.com with NNFMP; 23 Jan 2014 13:59:35 -0000
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rogers.com; s=s1024; t=1390485575; bh=d+M5Kr9pbR37o9lqr5FafeMilbi9dWILABcBmG2OIfU=; h=X-Yahoo-Newman-Id:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:X-Rocket-Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:CC:Subject:References:In-Reply-To:X-Enigmail-Version:Content-Type; b=Uq9pQe1F6tt6AhNUVHrfT/PTQV5Ijm6CCJh9fKP9v9vG7VQ32ycgC4A9jhk7FltrHb7BCXJfGxkfCMNBX9YyQulNVSXFYqwYHgWIDm98mpy1yTc8TnVQz4uMEY2X5zJLMseAgJ6Abf91zVPlpP5JDxZ5Tbj8c6KUTQSPRQo8ww8=
X-Yahoo-Newman-Id: 903449.94196.bm@smtp117.sbc.mail.gq1.yahoo.com
X-Yahoo-Newman-Property: ymail-3
X-YMail-OSG: xysnfJsVM1kpP4ePp7E1Crk.YFysw_DzaOvU0FGnXDbYkQf C.30_ynRhCazJ5gQYx2NWQHe4k2UVBbn0QpidV.d.bnscdf2w2WLjSUUr0mD X9kAzBGVqYnKzqZCem_CIWrLrjAGr6sFUjBGISpJqKhOkdn8MHEWlbzgo8i8 EloFoW8TztBLQQAqb3btFWL1r00mJoT7fW9wQTMtbr6hqYZY5n84ih5spLXF 6V4SHOtU2GatSakrYYLkkhWLS.24BY9Qfvu7uwQCvuQ9BMLwD0VS0JlNK9tC 4IRQmsIrfSgHpGBx_wytg71ls95DGr4CsYGV5g_GuJY6RODVpj491RvE8MYB yBF6ojk3nQK_.aWrnlat1Lz7gtuO0p0f.LYr1.e2OJ.X.sI_sFQEqWRRTUxq L0XEOMq1zs.ELg0.CtIhDvBmyl7QP_vx7Yy4X1arfXb97viSeWcJhkNyiZTx MQax1riZLSzCa89S8yhIGkoTbAR8s_tI2q3706pjuBg_B3Z2aKHpd8orOWbT qwVF5PO9TXJlZlCD9sdFk7vO_ufT.JVOLy_lMp2f.s1xrept3wOEBPW_K6RZ RDXTid9BZGx4WmEmLV5fH35PlaSjqtYDsVKPoGZ2_b6efqyGd.siBdp4FfGx 23_ZQD3eAf3rr34xaptIOvln54PrtQRlmDNRZ3U4nuw--
X-Yahoo-SMTP: sltvjZWswBCRD.ElTuB1l9j6s9wRYPpuyTNWOE5oEg--
X-Rocket-Received: from [10.111.100.153] (davec-b@74.213.188.46 with plain [67.195.22.104]) by smtp117.sbc.mail.gq1.yahoo.com with SMTP; 23 Jan 2014 13:59:35 +0000 UTC
Message-ID: <52E12045.9070705@rogers.com>
Date: Thu, 23 Jan 2014 08:59:33 -0500
From: David Collier-Brown <davec-b@rogers.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7
MIME-Version: 1.0
To: Dave Taht <dave.taht@gmail.com>
References: <52D9C1F3.3070803@rogers.com> <4FD254F4-D1E3-4CD1-AE35-C29C0DD37B4B@cisco.com> <CAA93jw4Yw=SHQYKMs5_LaRHYa18RYeKgo2gS5nET3XvhwR8Pjg@mail.gmail.com> <CAA93jw4Goh3Y4OrwTJDUcBGN0bj6YEAm++MYuAaqXiS4Er6q+g@mail.gmail.com>
In-Reply-To: <CAA93jw4Goh3Y4OrwTJDUcBGN0bj6YEAm++MYuAaqXiS4Er6q+g@mail.gmail.com>
X-Enigmail-Version: 1.6
Content-Type: multipart/alternative; boundary="------------010303050203080708040603"
Cc: "aqm@ietf.org" <aqm@ietf.org>, "Fred Baker \(fred\)" <fred@cisco.com>, "<davecb@spamcop.net>" <davecb@spamcop.net>
Subject: Re: [aqm] Prefatory comments re draft-aqm-reccommendation and -evaluation, and a question
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: davecb@spamcop.net
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 Jan 2014 13:59:40 -0000

This is a multi-part message in MIME format.
--------------010303050203080708040603
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit

[Inline]

On 01/23/2014 01:18 AM, Dave Taht wrote:
> On Thu, Jan 23, 2014 at 1:15 AM, Dave Taht <dave.taht@gmail.com> wrote:
>> On Thu, Jan 23, 2014 at 1:10 AM, Fred Baker (fred) <fred@cisco.com> wrote:
>>> No, you're not blowing smoke. I'm not sure I would compare the behavior to
>>> PMTUD, as in that the endpoint is given a magic number and manages to it,
>>> where in this case, it is given the results of its behavior, and it manages
>>> to improve that.
>>>
>>> But this is what I have rambled on about in threads relating to the size of
>>> a buffer. Folks would really like to have a magic way to calculate the
>>> buffer size (an amount of memory) they need to install in a router or
>>> switch, and it isn't that easy, because it has a lot to do with where in the
>>> network a system is located and how it is used by the applications that use
>>> it. But AQM, in the end, isn't about buffer size. It is about buffer
>>> occupancy. In the ideal case, if there are N sessions active on the
>>> bottleneck link in a path, we would like each to obtain 1/N of the
>>> bottleneck's capacity, which is to say that it should be able to maximize
>>> its throughput, while keeping an average of zero packets standing in the
>>> queue (minimizing both latency and variation in latency). If you know your
>>> math, you know that the ideal goal isn't actually achievable. But that
>>> doesn't stop us from trying to asymptotically approach it.
>> I prefer to think of the goal as to keep a minimum of 1 packet in the queue,
>> not as an average of 0.
> And that's not strictly true either. In the case of wifi, and other bundling
> technologies like those used in cable, you want to keep a minimum of a "good"
> aggregate of packets in the queue for that technology.

>From my pointy queue-head, occupancy is something that averages over
time, and has a good value if something less than 1.  However, that's
not necessarily one packet.

In  cases where packets are spaced out in time, an average of
0.<something> packets is what I expect.  Let's say 0.7 , arbitrarily.
When they come in trains, "bullet trains", or more likely, "thundering
herds", then 0.7 trains are what I expect to see.

The size of trains can be measured, but I expect the real thing we care
about is whether the queue drains back to zero during the time period
you're using for your measurements.  If it doesn't, you're about to be
in a world of hurt (:-))  It's only the physical size limitations of a
buffer that forces an eventual drop and congestion control to happen in
bad algorithms.  In the queuing world, we'd expect the queue to grow
"without bound", and the latency to try for infinity.

<start going off-topic>
Sun used to measure queue occupancy as a Riemann sum, which allowed one
to graph it like this:

1.0       
           ||
   ___   ||
  |    |   ||
  |    |   ||
  |    |   ||
_|    |__||__
0.0

The first one is 0.66 for 5 time units, the second 1.0 for one unit.
Both fall to zero, so the system catches up. They have different effects
on other traffic, which is why I find them interesting: as CPU queues,
the first hurts interactivity more than the second.

I *think* this will be of interest to us, but probably later.  I've
pasted in a chunk of the kstat.h comment at the very bottom for later
reference...
</off-topic>

Because occupancy can be done in units of time, we don't have to know
buffer and packet *sizes*

  * A well-performing queue/buffer system will have an average queue
    occupancy of < 1
  * A buffer is of sufficient size if the queue occupancy regularly
    falls to zero repeatedly during the time period. Another way of
    saying the same thing is that  the queue empties.
  * An operable queue management system sends congestion notifications
    if the queue/buffer does not empty
  * A good queue management system sends congestion notifications in
    time to keep the queue/buffer emptying regularly.

The parameter here is the time period, and has to do with the maximum
rate at which packets (work) can arrive. In other words, it's a
parameter based on the speed of the link.

--dave
[By the way, there are lots of better queue-heads than I.  The serious
mathies have way better intuition about what you're seeing that I do]

>
>>> On Jan 17, 2014, at 3:51 PM, David Collier-Brown <davec-b@rogers.com> wrote:
>>>
>>> I've been reading through the internet-drafts, and one paragraph struck me
>>> as very illuminating.  This is therefor a sanity-check before I go full-hog
>>> down a particular path...
>>>
>>> The comment is from Baker and Fairhurst,
>>> https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/ and the
>>> paragraph is [emphases added]
>>>
>>> The point of buffering in the network is to absorb data bursts and to
>>> transmit them during the (hopefully) ensuing bursts of silence. This is
>>> essential to permit the transmission of bursty data. Normally small queues
>>> are preferred in network devices, with sufficient queue capacity to absorb
>>> the bursts. The counter-intuitive result is that maintaining normally-small
>>> queues can result in higher throughput as well as lower end-to- end delay.
>>> In summary, queue limits should not reflect the steady state queues we want
>>> to be maintained in the network; instead, they should reflect the size of
>>> bursts that a network device needs to absorb.
>>>
>>> All of a sudden we're talking about the kinds of queues I know a little
>>> about (:-))
>>> ---
>>>
>>> I'm going to suggest that these are queues and associated physical buffers
>>> that do two things:
>>>
>>> hold packets that arrive at a bottleneck for a long as it takes to send them
>>> out a slower link that they came in on, and
>>> hold bursts of packets that arrive adjacent to each other until they can be
>>> sent out in a normal spacing, with some small amount of time between them
>>>
>>>
>>> In an illustration of Dave Taht's, the first looks something like this
>>>
>>> -------------------------+
>>>     |X|            |X|      +-------------------
>>>     |X|            |X|         |XXX|    |XXX|
>>>     |X|            |X|      +-------------------
>>> -------------------------+
>>>
>>> At the choke-point there is a buffer at least big enough to give the packet
>>> a chance to wheel from line into column (:-)) and start down the smaller
>>> pipe.
>>>
>>> The speed at which the acks come back,  the frequency of drops, and any
>>> explicit congestion notifications slows the sender until they don't overload
>>> the skinnier pipe, thus spacing the packets in the fatter pipe out.
>>>
>>> Various causes [Leland] can slow or speed the packets in the fat pipe,
>>> making it possible for several to arrive adjacent to each other, followed by
>>> a gap.  The second purpose of a buffer is to hold  these bursts while things
>>> space themselves back out.
>>>
>>> They need to be big enough at minimum to do the speed matching,  and at
>>> maximum,  big enough to spread a burst back into a normal progression,
>>> always assuming that acks, drops and explicit congestion notifications are
>>> slowing the sender to the speed of the slowest part of the network.
>>> ---
>>>
>>> If I'm right about this, we can draw some helpful conclusions
>>>
>>> buffer sizes can be set based on measurements:
>>>
>>> speed differences, which are pretty static, plus
>>> observed burstyness
>>>
>>> drops and ECN can be done to match the slowest speed in the path
>>>
>>> The latter suddenly sounds a bit like path MTU discovery, except it's a bit
>>> more dynamic, and varies with both path and what's happening in various
>>> parts of it.
>>>
>>> To me, as a capacity/performance nerd, this sounds a lot more familiar and
>>> manageable. My question to you, before I start madly scribbling on the
>>> internet drafts is:
>>>
>>>     Am I blowing smoke?
>>>
>>> --dave
>>>
>>> --
>>> David Collier-Brown,         | Always do right. This will gratify
>>> System Programmer and Author | some people and astonish the rest
>>> davecb@spamcop.net           |                      -- Mark Twain
>>> (416) 223-8968
>>>
>>> _______________________________________________
>>>
>>> aqm mailing list
>>> aqm@ietf.org
>>> https://www.ietf.org/mailman/listinfo/aqm
>>>
>>>
>>>
>>
>>
>> --
>> Dave Täht
>>
>> Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html
>
>

       /*
         * Accumulated time and queue length statistics.
         *
         * Accumulated time statistics are kept as a running sum
         * of "active" time.  Queue length statistics are kept as a
         * running sum of the product of queue length and elapsed time
         * at that length -- i.e., a Riemann sum for queue length
         * integrated against time.  (You can also think of the active time
         * as a Riemann sum, for the boolean function (queue_length > 0)
         * integrated against time, or you can think of it as the
         * Lebesgue measure of the set on which queue_length > 0.)
         *
         *              ^
         *              |                       _________
         *              8                       | i4    |
         *              |                       |       |
         *      Queue   6                       |       |
         *      Length  |       _________       |       |
         *              4       | i2    |_______|       |
         *              |       |           i3          |
         *              2_______|                       |
         *              |    i1                         |
         *              |_______________________________|
         *              Time->  t1      t2      t3      t4
         * At each change of state (entry or exit from the queue),
         * we add the elapsed time (since the previous state change)
         * to the active time if the queue length was non-zero during
         * that interval; and we add the product of the elapsed time
         * times the queue length to the running length*time sum.
         *
         * This method is generalizable to measuring residency
         * in any defined system: instead of queue lengths, think
         * of "outstanding RPC calls to server X".
         *
         * A large number of I/O subsystems have at least two basic
         * "lists" of transactions they manage: one for transactions
         * that have been accepted for processing but for which processing
         * has yet to begin, and one for transactions which are actively
         * being processed (but not done). For this reason, two cumulative
         * time statistics are defined here: wait (pre-service) time,
         * and run (service) time.
         *
         * All times are 64-bit nanoseconds (hrtime_t), as returned by
         * gethrtime().
         *
         * The units of cumulative busy time are accumulated nanoseconds.
         * The units of cumulative length*time products are elapsed time
         * times queue length.        
         */


The ascii art is an excerpt from /usr/include/sys/kstat.h on Solaris 10



-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
davecb@spamcop.net           |                      -- Mark Twain
(416) 223-8968


--------------010303050203080708040603
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">[Inline]<br>
      <br>
      On 01/23/2014 01:18 AM, Dave Taht wrote:<br>
    </div>
    <blockquote
cite="mid:CAA93jw4Goh3Y4OrwTJDUcBGN0bj6YEAm++MYuAaqXiS4Er6q+g@mail.gmail.com"
      type="cite">
      <pre wrap="">On Thu, Jan 23, 2014 at 1:15 AM, Dave Taht <a class="moz-txt-link-rfc2396E" href="mailto:dave.taht@gmail.com">&lt;dave.taht@gmail.com&gt;</a> wrote:
</pre>
      <blockquote type="cite">
        <pre wrap="">On Thu, Jan 23, 2014 at 1:10 AM, Fred Baker (fred) <a class="moz-txt-link-rfc2396E" href="mailto:fred@cisco.com">&lt;fred@cisco.com&gt;</a> wrote:
</pre>
        <blockquote type="cite">
          <pre wrap="">No, you're not blowing smoke. I'm not sure I would compare the behavior to
PMTUD, as in that the endpoint is given a magic number and manages to it,
where in this case, it is given the results of its behavior, and it manages
to improve that.

But this is what I have rambled on about in threads relating to the size of
a buffer. Folks would really like to have a magic way to calculate the
buffer size (an amount of memory) they need to install in a router or
switch, and it isn't that easy, because it has a lot to do with where in the
network a system is located and how it is used by the applications that use
it. But AQM, in the end, isn't about buffer size. It is about buffer
occupancy. In the ideal case, if there are N sessions active on the
bottleneck link in a path, we would like each to obtain 1/N of the
bottleneck's capacity, which is to say that it should be able to maximize
its throughput, while keeping an average of zero packets standing in the
queue (minimizing both latency and variation in latency). If you know your
math, you know that the ideal goal isn't actually achievable. But that
doesn't stop us from trying to asymptotically approach it.
</pre>
        </blockquote>
        <pre wrap="">
I prefer to think of the goal as to keep a minimum of 1 packet in the queue,
not as an average of 0.
</pre>
      </blockquote>
      <pre wrap="">
And that's not strictly true either. In the case of wifi, and other bundling
technologies like those used in cable, you want to keep a minimum of a "good"
aggregate of packets in the queue for that technology.</pre>
    </blockquote>
    <br>
    From my pointy queue-head, occupancy is something that averages over
    time, and has a good value if something less than 1.&nbsp; However,
    that's not necessarily one packet.<br>
    <br>
    In&nbsp; cases where packets are spaced out in time, an average of
    0.&lt;something&gt; packets is what I expect.&nbsp; Let's say 0.7 ,
    arbitrarily. When they come in trains, "bullet trains", or more
    likely, "thundering herds", then 0.7 trains are what I expect to
    see.<br>
    <br>
    The size of trains can be measured, but I expect the real thing we
    care about is whether the queue drains back to zero during the time
    period you're using for your measurements.&nbsp; If it doesn't, you're
    about to be in a world of hurt (:-))&nbsp; It's only the physical size
    limitations of a buffer that forces an eventual drop and congestion
    control to happen in bad algorithms.&nbsp; In the queuing world, we'd
    expect the queue to grow "without bound", and the latency to try for
    infinity.<br>
    <br>
    &lt;start going off-topic&gt;<br>
    Sun used to measure queue occupancy as a Riemann sum, which allowed
    one to graph it like this:<br>
    <br>
    1.0 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <br>
    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; ||<br>
    &nbsp;&nbsp; ___&nbsp;&nbsp; ||<br>
    &nbsp; | &nbsp;&nbsp; | &nbsp; ||<br>
    &nbsp; |&nbsp;&nbsp;&nbsp; |&nbsp;&nbsp; ||<br>
    &nbsp; |&nbsp;&nbsp;&nbsp; |&nbsp;&nbsp; ||<br>
    _|&nbsp;&nbsp;&nbsp; |__||__<br>
    0.0<br>
    <br>
    The first one is 0.66 for 5 time units, the second 1.0 for one unit.<br>
    Both fall to zero, so the system catches up. They have different
    effects on other traffic, which is why I find them interesting: as
    CPU queues, the first hurts interactivity more than the second.<br>
    <br>
    I *think* this will be of interest to us, but probably later.&nbsp; I've
    pasted in a chunk of the kstat.h comment at the very bottom for
    later reference...<br>
    &lt;/off-topic&gt;<br>
    <br>
    Because occupancy can be done in units of time, we don't have to
    know buffer and packet *sizes*<br>
    <ul>
      <li>A well-performing queue/buffer system will have an average
        queue occupancy of &lt; 1</li>
      <li>A buffer is of sufficient size if the queue occupancy
        regularly falls to zero repeatedly during the time period.
        Another way of saying the same thing is that&nbsp; the queue empties.</li>
      <li>An operable queue management system sends congestion
        notifications if the queue/buffer does not empty</li>
      <li>A good queue management system sends congestion notifications
        in time to keep the queue/buffer emptying regularly.</li>
    </ul>
    The parameter here is the time period, and has to do with the
    maximum rate at which packets (work) can arrive. In other words,
    it's a parameter based on the speed of the link.<br>
    <br>
    --dave<br>
    [By the way, there are lots of better queue-heads than I.&nbsp; The
    serious mathies have way better intuition about what you're seeing
    that I do]<br>
    <br>
    <blockquote
cite="mid:CAA93jw4Goh3Y4OrwTJDUcBGN0bj6YEAm++MYuAaqXiS4Er6q+g@mail.gmail.com"
      type="cite">
      <pre wrap="">

</pre>
      <blockquote type="cite">
        <pre wrap="">
</pre>
        <blockquote type="cite">
          <pre wrap="">
On Jan 17, 2014, at 3:51 PM, David Collier-Brown <a class="moz-txt-link-rfc2396E" href="mailto:davec-b@rogers.com">&lt;davec-b@rogers.com&gt;</a> wrote:

I've been reading through the internet-drafts, and one paragraph struck me
as very illuminating.  This is therefor a sanity-check before I go full-hog
down a particular path...

The comment is from Baker and Fairhurst,
<a class="moz-txt-link-freetext" href="https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/">https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/</a> and the
paragraph is [emphases added]

The point of buffering in the network is to absorb data bursts and to
transmit them during the (hopefully) ensuing bursts of silence. This is
essential to permit the transmission of bursty data. Normally small queues
are preferred in network devices, with sufficient queue capacity to absorb
the bursts. The counter-intuitive result is that maintaining normally-small
queues can result in higher throughput as well as lower end-to- end delay.
In summary, queue limits should not reflect the steady state queues we want
to be maintained in the network; instead, they should reflect the size of
bursts that a network device needs to absorb.

All of a sudden we're talking about the kinds of queues I know a little
about (:-))
---

I'm going to suggest that these are queues and associated physical buffers
that do two things:

hold packets that arrive at a bottleneck for a long as it takes to send them
out a slower link that they came in on, and
hold bursts of packets that arrive adjacent to each other until they can be
sent out in a normal spacing, with some small amount of time between them


In an illustration of Dave Taht's, the first looks something like this

-------------------------+
    |X|            |X|      +-------------------
    |X|            |X|         |XXX|    |XXX|
    |X|            |X|      +-------------------
-------------------------+

At the choke-point there is a buffer at least big enough to give the packet
a chance to wheel from line into column (:-)) and start down the smaller
pipe.

The speed at which the acks come back,  the frequency of drops, and any
explicit congestion notifications slows the sender until they don't overload
the skinnier pipe, thus spacing the packets in the fatter pipe out.

Various causes [Leland] can slow or speed the packets in the fat pipe,
making it possible for several to arrive adjacent to each other, followed by
a gap.  The second purpose of a buffer is to hold  these bursts while things
space themselves back out.

They need to be big enough at minimum to do the speed matching,  and at
maximum,  big enough to spread a burst back into a normal progression,
always assuming that acks, drops and explicit congestion notifications are
slowing the sender to the speed of the slowest part of the network.
---

If I'm right about this, we can draw some helpful conclusions

buffer sizes can be set based on measurements:

speed differences, which are pretty static, plus
observed burstyness

drops and ECN can be done to match the slowest speed in the path

The latter suddenly sounds a bit like path MTU discovery, except it's a bit
more dynamic, and varies with both path and what's happening in various
parts of it.

To me, as a capacity/performance nerd, this sounds a lot more familiar and
manageable. My question to you, before I start madly scribbling on the
internet drafts is:

    Am I blowing smoke?

--dave

--
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
<a class="moz-txt-link-abbreviated" href="mailto:davecb@spamcop.net">davecb@spamcop.net</a>           |                      -- Mark Twain
(416) 223-8968

_______________________________________________

aqm mailing list
<a class="moz-txt-link-abbreviated" href="mailto:aqm@ietf.org">aqm@ietf.org</a>
<a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/aqm">https://www.ietf.org/mailman/listinfo/aqm</a>



</pre>
        </blockquote>
        <pre wrap="">


--
Dave T&auml;ht

Fixing bufferbloat with cerowrt: <a class="moz-txt-link-freetext" href="http://www.teklibre.com/cerowrt/subscribe.html">http://www.teklibre.com/cerowrt/subscribe.html</a>
</pre>
      </blockquote>
      <pre wrap="">


</pre>
    </blockquote>
    <br>
    <pre wrap="">       /*
         * Accumulated time and queue length statistics.
         *
         * Accumulated time statistics are kept as a running sum
         * of "active" time.  Queue length statistics are kept as a
         * running sum of the product of queue length and elapsed time
         * at that length -- i.e., a Riemann sum for queue length
         * integrated against time.  (You can also think of the active time
         * as a Riemann sum, for the boolean function (queue_length &gt; 0)
         * integrated against time, or you can think of it as the
         * Lebesgue measure of the set on which queue_length &gt; 0.)
         *
         *              ^
         *              |                       _________
         *              8                       | i4    |
         *              |                       |       |
         *      Queue   6                       |       |
         *      Length  |       _________       |       |
         *              4       | i2    |_______|       |
         *              |       |           i3          |
         *              2_______|                       |
         *              |    i1                         |
         *              |_______________________________|
         *              Time-&gt;  t1      t2      t3      t4
         * At each change of state (entry or exit from the queue),
         * we add the elapsed time (since the previous state change)
         * to the active time if the queue length was non-zero during
         * that interval; and we add the product of the elapsed time
         * times the queue length to the running length*time sum.
         *
         * This method is generalizable to measuring residency
         * in any defined system: instead of queue lengths, think
         * of "outstanding RPC calls to server X".
         *
         * A large number of I/O subsystems have at least two basic
         * "lists" of transactions they manage: one for transactions
         * that have been accepted for processing but for which processing
         * has yet to begin, and one for transactions which are actively
         * being processed (but not done). For this reason, two cumulative
         * time statistics are defined here: wait (pre-service) time,
         * and run (service) time.
         *
         * All times are 64-bit nanoseconds (hrtime_t), as returned by
         * gethrtime().
         *
         * The units of cumulative busy time are accumulated nanoseconds.
         * The units of cumulative length*time products are elapsed time
         * times queue length.        
         */


The ascii art is an excerpt from /usr/include/sys/kstat.h on Solaris 10</pre>
    <br>
    <br>
    <pre class="moz-signature" cols="72">-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
<a class="moz-txt-link-abbreviated" href="mailto:davecb@spamcop.net">davecb@spamcop.net</a>           |                      -- Mark Twain
(416) 223-8968
</pre>
  </body>
</html>

--------------010303050203080708040603--

From davec-b@rogers.com  Thu Jan 23 06:42:34 2014
Return-Path: <davec-b@rogers.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E4AF81A0461 for <aqm@ietfa.amsl.com>; Thu, 23 Jan 2014 06:42:34 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.299
X-Spam-Level: 
X-Spam-Status: No, score=-1.299 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, HTML_MESSAGE=0.001, J_CHICKENPOX_64=0.6, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RAY_Q-YRm5ff for <aqm@ietfa.amsl.com>; Thu, 23 Jan 2014 06:42:31 -0800 (PST)
Received: from nm8-vm2.access.bullet.mail.bf1.yahoo.com (nm8-vm2.access.bullet.mail.bf1.yahoo.com [216.109.114.177]) by ietfa.amsl.com (Postfix) with ESMTP id 9564D1A045D for <aqm@ietf.org>; Thu, 23 Jan 2014 06:42:31 -0800 (PST)
Received: from [66.196.81.165] by nm8.access.bullet.mail.bf1.yahoo.com with NNFMP; 23 Jan 2014 14:42:30 -0000
Received: from [98.139.244.49] by tm11.access.bullet.mail.bf1.yahoo.com with NNFMP; 23 Jan 2014 14:42:30 -0000
Received: from [127.0.0.1] by smtp111.sbc.mail.bf1.yahoo.com with NNFMP; 23 Jan 2014 14:42:30 -0000
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1390488150; bh=fGeSxYxvZ4Kla8w3eWJmhiejdKJscijmGhc0q9doStc=; h=X-Yahoo-Newman-Id:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:X-Rocket-Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:Subject:References:In-Reply-To:X-Enigmail-Version:Content-Type; b=FuHLvk6+Wa9D61F8ByR63xLYGJs3w0+AP4hYw6XfBXx3Y6SCc1/xzVPMIGcSsMqtCPLcISzt+5w1rEw1xT0iRwmJOUyZTFAqxfDuRQxhdzVhX5VIOpMequiYb2Jvt+84NjhTDNpjB2lErXL9RvMLAnxFvRHKy59wvWfzmZ7Tq2E=
X-Yahoo-Newman-Id: 483912.65392.bm@smtp111.sbc.mail.bf1.yahoo.com
X-Yahoo-Newman-Property: ymail-3
X-YMail-OSG: nKyeaDwVM1lG4z.MbhVtePwNkbyzQLN926HWKH_nj63U079 shrmL5u2PqfaaTHVIaU1J9j0s1uDf2_xvrPCswftaUwxWIwjecNuQjJeEIwn dn2u6n7rL5FErCsInJHttxVbiFLJl1KoSDilLk2d0ia3jh212PwOI7LvgFka pOU9N5EPpuIcKe7bVvpPNfmpd14YuwR3.BTgFuaMQPbFuD960CafT.beeEjg QsiQNAjZsIaS8PXOdRgNGPtcxDvyAFQuKOQgSYzUWUAuNJC55C6VAVmLjhPr hP2K_az2zvhzokDjp5maxP90XS4kakXw8E4tsnrlJRSaZvMF2BEDFfyCW.BO PbUznLGdugx0Ja4l0OSz50Xi007Sm7wInL6viwxfpRu_8kTHKdlkE7uRM_nm hhWpjfj1me1MqRkUfkkCTWRVKRnrQBQmLSWCdrDH.r_pYiEmJuS95qYdFGnr J8L_XFa2ZJKcwMjbRXGH.a82eyd76_pzK82cBhgIbK2J.qroultrCsHtxM2j xrQJiNvPGk1D0177.v2JzJ4UkLD6cCTM7UFSmvMPnF3wLsX9DC_nsM5wQnTA DYRLEk._fNYdOmHENaDYMhzsXTufjfHCvR6y32agN5ihEZmYpbtMfEZdf5dG QAmL0rNREuAjEpNCSzfIse02bXQ5gXql6nhCs
X-Yahoo-SMTP: sltvjZWswBCRD.ElTuB1l9j6s9wRYPpuyTNWOE5oEg--
X-Rocket-Received: from [10.111.100.153] (davec-b@74.213.188.46 with plain [98.139.221.125]) by smtp111.sbc.mail.bf1.yahoo.com with SMTP; 23 Jan 2014 06:42:30 -0800 PST
Message-ID: <52E12A55.7080401@rogers.com>
Date: Thu, 23 Jan 2014 09:42:29 -0500
From: David Collier-Brown <davec-b@rogers.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7
MIME-Version: 1.0
To: aqm@ietf.org
References: <52D9C1F3.3070803@rogers.com> <4FD254F4-D1E3-4CD1-AE35-C29C0DD37B4B@cisco.com> <CAA93jw4Yw=SHQYKMs5_LaRHYa18RYeKgo2gS5nET3XvhwR8Pjg@mail.gmail.com> <CAA93jw4Goh3Y4OrwTJDUcBGN0bj6YEAm++MYuAaqXiS4Er6q+g@mail.gmail.com> <52E12045.9070705@rogers.com>
In-Reply-To: <52E12045.9070705@rogers.com>
X-Enigmail-Version: 1.6
Content-Type: multipart/alternative; boundary="------------070508030001090201030906"
Subject: Re: [aqm] Prefatory comments re draft-aqm-reccommendation and -evaluation, and a question
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: davecb@spamcop.net
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 Jan 2014 14:42:35 -0000

This is a multi-part message in MIME format.
--------------070508030001090201030906
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit

Addressing buffer size when designing hardware, it would be an economic
discussion illuminated by data from the last model of my device that had
a good AQM solution.  With AQM in place, buffers would only overflow
when things got seriously bursty, and some data* from the devices in the
field would tell me if I wanted to go to the next larger memory chip.

--dave
[* suitably anonymized]

On 01/23/2014 08:59 AM, David Collier-Brown wrote:
> [Inline]
>
> On 01/23/2014 01:18 AM, Dave Taht wrote:
>> On Thu, Jan 23, 2014 at 1:15 AM, Dave Taht <dave.taht@gmail.com> wrote:
>>> On Thu, Jan 23, 2014 at 1:10 AM, Fred Baker (fred) <fred@cisco.com> wrote:
>>>> No, you're not blowing smoke. I'm not sure I would compare the behavior to
>>>> PMTUD, as in that the endpoint is given a magic number and manages to it,
>>>> where in this case, it is given the results of its behavior, and it manages
>>>> to improve that.
>>>>
>>>> But this is what I have rambled on about in threads relating to the size of
>>>> a buffer. Folks would really like to have a magic way to calculate the
>>>> buffer size (an amount of memory) they need to install in a router or
>>>> switch, and it isn't that easy, because it has a lot to do with where in the
>>>> network a system is located and how it is used by the applications that use
>>>> it. But AQM, in the end, isn't about buffer size. It is about buffer
>>>> occupancy. In the ideal case, if there are N sessions active on the
>>>> bottleneck link in a path, we would like each to obtain 1/N of the
>>>> bottleneck's capacity, which is to say that it should be able to maximize
>>>> its throughput, while keeping an average of zero packets standing in the
>>>> queue (minimizing both latency and variation in latency). If you know your
>>>> math, you know that the ideal goal isn't actually achievable. But that
>>>> doesn't stop us from trying to asymptotically approach it.
>>> I prefer to think of the goal as to keep a minimum of 1 packet in the queue,
>>> not as an average of 0.
>> And that's not strictly true either. In the case of wifi, and other bundling
>> technologies like those used in cable, you want to keep a minimum of a "good"
>> aggregate of packets in the queue for that technology.
>
> From my pointy queue-head, occupancy is something that averages over
> time, and has a good value if something less than 1.  However, that's
> not necessarily one packet.
>
> In  cases where packets are spaced out in time, an average of
> 0.<something> packets is what I expect.  Let's say 0.7 , arbitrarily.
> When they come in trains, "bullet trains", or more likely, "thundering
> herds", then 0.7 trains are what I expect to see.
>
> The size of trains can be measured, but I expect the real thing we
> care about is whether the queue drains back to zero during the time
> period you're using for your measurements.  If it doesn't, you're
> about to be in a world of hurt (:-))  It's only the physical size
> limitations of a buffer that forces an eventual drop and congestion
> control to happen in bad algorithms.  In the queuing world, we'd
> expect the queue to grow "without bound", and the latency to try for
> infinity.
>
> <start going off-topic>
> Sun used to measure queue occupancy as a Riemann sum, which allowed
> one to graph it like this:
>
> 1.0       
>            ||
>    ___   ||
>   |    |   ||
>   |    |   ||
>   |    |   ||
> _|    |__||__
> 0.0
>
> The first one is 0.66 for 5 time units, the second 1.0 for one unit.
> Both fall to zero, so the system catches up. They have different
> effects on other traffic, which is why I find them interesting: as CPU
> queues, the first hurts interactivity more than the second.
>
> I *think* this will be of interest to us, but probably later.  I've
> pasted in a chunk of the kstat.h comment at the very bottom for later
> reference...
> </off-topic>
>
> Because occupancy can be done in units of time, we don't have to know
> buffer and packet *sizes*
>
>   * A well-performing queue/buffer system will have an average queue
>     occupancy of < 1
>   * A buffer is of sufficient size if the queue occupancy regularly
>     falls to zero repeatedly during the time period. Another way of
>     saying the same thing is that  the queue empties.
>   * An operable queue management system sends congestion notifications
>     if the queue/buffer does not empty
>   * A good queue management system sends congestion notifications in
>     time to keep the queue/buffer emptying regularly.
>
> The parameter here is the time period, and has to do with the maximum
> rate at which packets (work) can arrive. In other words, it's a
> parameter based on the speed of the link.
>
> --dave
> [By the way, there are lots of better queue-heads than I.  The serious
> mathies have way better intuition about what you're seeing that I do]
>
>>>> On Jan 17, 2014, at 3:51 PM, David Collier-Brown <davec-b@rogers.com> wrote:
>>>>
>>>> I've been reading through the internet-drafts, and one paragraph struck me
>>>> as very illuminating.  This is therefor a sanity-check before I go full-hog
>>>> down a particular path...
>>>>
>>>> The comment is from Baker and Fairhurst,
>>>> https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/ and the
>>>> paragraph is [emphases added]
>>>>
>>>> The point of buffering in the network is to absorb data bursts and to
>>>> transmit them during the (hopefully) ensuing bursts of silence. This is
>>>> essential to permit the transmission of bursty data. Normally small queues
>>>> are preferred in network devices, with sufficient queue capacity to absorb
>>>> the bursts. The counter-intuitive result is that maintaining normally-small
>>>> queues can result in higher throughput as well as lower end-to- end delay.
>>>> In summary, queue limits should not reflect the steady state queues we want
>>>> to be maintained in the network; instead, they should reflect the size of
>>>> bursts that a network device needs to absorb.
>>>>
>>>> All of a sudden we're talking about the kinds of queues I know a little
>>>> about (:-))
>>>> ---
>>>>
>>>> I'm going to suggest that these are queues and associated physical buffers
>>>> that do two things:
>>>>
>>>> hold packets that arrive at a bottleneck for a long as it takes to send them
>>>> out a slower link that they came in on, and
>>>> hold bursts of packets that arrive adjacent to each other until they can be
>>>> sent out in a normal spacing, with some small amount of time between them
>>>>
>>>>
>>>> In an illustration of Dave Taht's, the first looks something like this
>>>>
>>>> -------------------------+
>>>>     |X|            |X|      +-------------------
>>>>     |X|            |X|         |XXX|    |XXX|
>>>>     |X|            |X|      +-------------------
>>>> -------------------------+
>>>>
>>>> At the choke-point there is a buffer at least big enough to give the packet
>>>> a chance to wheel from line into column (:-)) and start down the smaller
>>>> pipe.
>>>>
>>>> The speed at which the acks come back,  the frequency of drops, and any
>>>> explicit congestion notifications slows the sender until they don't overload
>>>> the skinnier pipe, thus spacing the packets in the fatter pipe out.
>>>>
>>>> Various causes [Leland] can slow or speed the packets in the fat pipe,
>>>> making it possible for several to arrive adjacent to each other, followed by
>>>> a gap.  The second purpose of a buffer is to hold  these bursts while things
>>>> space themselves back out.
>>>>
>>>> They need to be big enough at minimum to do the speed matching,  and at
>>>> maximum,  big enough to spread a burst back into a normal progression,
>>>> always assuming that acks, drops and explicit congestion notifications are
>>>> slowing the sender to the speed of the slowest part of the network.
>>>> ---
>>>>
>>>> If I'm right about this, we can draw some helpful conclusions
>>>>
>>>> buffer sizes can be set based on measurements:
>>>>
>>>> speed differences, which are pretty static, plus
>>>> observed burstyness
>>>>
>>>> drops and ECN can be done to match the slowest speed in the path
>>>>
>>>> The latter suddenly sounds a bit like path MTU discovery, except it's a bit
>>>> more dynamic, and varies with both path and what's happening in various
>>>> parts of it.
>>>>
>>>> To me, as a capacity/performance nerd, this sounds a lot more familiar and
>>>> manageable. My question to you, before I start madly scribbling on the
>>>> internet drafts is:
>>>>
>>>>     Am I blowing smoke?
>>>>
>>>> --dave
>>>>
>>>> --
>>>> David Collier-Brown,         | Always do right. This will gratify
>>>> System Programmer and Author | some people and astonish the rest
>>>> davecb@spamcop.net           |                      -- Mark Twain
>>>> (416) 223-8968
>>>>
>>>> _______________________________________________
>>>>
>>>> aqm mailing list
>>>> aqm@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/aqm
>>>>
>>>>
>>>>
>>>
>>> --
>>> Dave Täht
>>>
>>> Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html
>>
>
>        /*
>          * Accumulated time and queue length statistics.
>          *
>          * Accumulated time statistics are kept as a running sum
>          * of "active" time.  Queue length statistics are kept as a
>          * running sum of the product of queue length and elapsed time
>          * at that length -- i.e., a Riemann sum for queue length
>          * integrated against time.  (You can also think of the active time
>          * as a Riemann sum, for the boolean function (queue_length > 0)
>          * integrated against time, or you can think of it as the
>          * Lebesgue measure of the set on which queue_length > 0.)
>          *
>          *              ^
>          *              |                       _________
>          *              8                       | i4    |
>          *              |                       |       |
>          *      Queue   6                       |       |
>          *      Length  |       _________       |       |
>          *              4       | i2    |_______|       |
>          *              |       |           i3          |
>          *              2_______|                       |
>          *              |    i1                         |
>          *              |_______________________________|
>          *              Time->  t1      t2      t3      t4
>          * At each change of state (entry or exit from the queue),
>          * we add the elapsed time (since the previous state change)
>          * to the active time if the queue length was non-zero during
>          * that interval; and we add the product of the elapsed time
>          * times the queue length to the running length*time sum.
>          *
>          * This method is generalizable to measuring residency
>          * in any defined system: instead of queue lengths, think
>          * of "outstanding RPC calls to server X".
>          *
>          * A large number of I/O subsystems have at least two basic
>          * "lists" of transactions they manage: one for transactions
>          * that have been accepted for processing but for which processing
>          * has yet to begin, and one for transactions which are actively
>          * being processed (but not done). For this reason, two cumulative
>          * time statistics are defined here: wait (pre-service) time,
>          * and run (service) time.
>          *
>          * All times are 64-bit nanoseconds (hrtime_t), as returned by
>          * gethrtime().
>          *
>          * The units of cumulative busy time are accumulated nanoseconds.
>          * The units of cumulative length*time products are elapsed time
>          * times queue length.        
>          */
>
>
> The ascii art is an excerpt from /usr/include/sys/kstat.h on Solaris 10
>
>
> -- 
> David Collier-Brown,         | Always do right. This will gratify
> System Programmer and Author | some people and astonish the rest
> davecb@spamcop.net           |                      -- Mark Twain
> (416) 223-8968
>
>
> _______________________________________________
> aqm mailing list
> aqm@ietf.org
> https://www.ietf.org/mailman/listinfo/aqm


-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
davecb@spamcop.net           |                      -- Mark Twain
(416) 223-8968


--------------070508030001090201030906
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">Addressing buffer size when designing
      hardware, it would be an economic discussion illuminated by data
      from the last model of my device that had a good AQM solution.&nbsp;
      With AQM in place, buffers would only overflow when things got
      seriously bursty, and some data* from the devices in the field
      would tell me if I wanted to go to the next larger memory chip.<br>
      <br>
      --dave<br>
      [* suitably anonymized]<br>
      <br>
      On 01/23/2014 08:59 AM, David Collier-Brown wrote:<br>
    </div>
    <blockquote cite="mid:52E12045.9070705@rogers.com" type="cite">
      <meta content="text/html; charset=ISO-8859-1"
        http-equiv="Content-Type">
      <div class="moz-cite-prefix">[Inline]<br>
        <br>
        On 01/23/2014 01:18 AM, Dave Taht wrote:<br>
      </div>
      <blockquote
cite="mid:CAA93jw4Goh3Y4OrwTJDUcBGN0bj6YEAm++MYuAaqXiS4Er6q+g@mail.gmail.com"
        type="cite">
        <pre wrap="">On Thu, Jan 23, 2014 at 1:15 AM, Dave Taht <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:dave.taht@gmail.com">&lt;dave.taht@gmail.com&gt;</a> wrote:
</pre>
        <blockquote type="cite">
          <pre wrap="">On Thu, Jan 23, 2014 at 1:10 AM, Fred Baker (fred) <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:fred@cisco.com">&lt;fred@cisco.com&gt;</a> wrote:
</pre>
          <blockquote type="cite">
            <pre wrap="">No, you're not blowing smoke. I'm not sure I would compare the behavior to
PMTUD, as in that the endpoint is given a magic number and manages to it,
where in this case, it is given the results of its behavior, and it manages
to improve that.

But this is what I have rambled on about in threads relating to the size of
a buffer. Folks would really like to have a magic way to calculate the
buffer size (an amount of memory) they need to install in a router or
switch, and it isn't that easy, because it has a lot to do with where in the
network a system is located and how it is used by the applications that use
it. But AQM, in the end, isn't about buffer size. It is about buffer
occupancy. In the ideal case, if there are N sessions active on the
bottleneck link in a path, we would like each to obtain 1/N of the
bottleneck's capacity, which is to say that it should be able to maximize
its throughput, while keeping an average of zero packets standing in the
queue (minimizing both latency and variation in latency). If you know your
math, you know that the ideal goal isn't actually achievable. But that
doesn't stop us from trying to asymptotically approach it.
</pre>
          </blockquote>
          <pre wrap="">I prefer to think of the goal as to keep a minimum of 1 packet in the queue,
not as an average of 0.
</pre>
        </blockquote>
        <pre wrap="">And that's not strictly true either. In the case of wifi, and other bundling
technologies like those used in cable, you want to keep a minimum of a "good"
aggregate of packets in the queue for that technology.</pre>
      </blockquote>
      <br>
      From my pointy queue-head, occupancy is something that averages
      over time, and has a good value if something less than 1.&nbsp;
      However, that's not necessarily one packet.<br>
      <br>
      In&nbsp; cases where packets are spaced out in time, an average of
      0.&lt;something&gt; packets is what I expect.&nbsp; Let's say 0.7 ,
      arbitrarily. When they come in trains, "bullet trains", or more
      likely, "thundering herds", then 0.7 trains are what I expect to
      see.<br>
      <br>
      The size of trains can be measured, but I expect the real thing we
      care about is whether the queue drains back to zero during the
      time period you're using for your measurements.&nbsp; If it doesn't,
      you're about to be in a world of hurt (:-))&nbsp; It's only the
      physical size limitations of a buffer that forces an eventual drop
      and congestion control to happen in bad algorithms.&nbsp; In the
      queuing world, we'd expect the queue to grow "without bound", and
      the latency to try for infinity.<br>
      <br>
      &lt;start going off-topic&gt;<br>
      Sun used to measure queue occupancy as a Riemann sum, which
      allowed one to graph it like this:<br>
      <br>
      1.0 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <br>
      &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; ||<br>
      &nbsp;&nbsp; ___&nbsp;&nbsp; ||<br>
      &nbsp; | &nbsp;&nbsp; | &nbsp; ||<br>
      &nbsp; |&nbsp;&nbsp;&nbsp; |&nbsp;&nbsp; ||<br>
      &nbsp; |&nbsp;&nbsp;&nbsp; |&nbsp;&nbsp; ||<br>
      _|&nbsp;&nbsp;&nbsp; |__||__<br>
      0.0<br>
      <br>
      The first one is 0.66 for 5 time units, the second 1.0 for one
      unit.<br>
      Both fall to zero, so the system catches up. They have different
      effects on other traffic, which is why I find them interesting: as
      CPU queues, the first hurts interactivity more than the second.<br>
      <br>
      I *think* this will be of interest to us, but probably later.&nbsp;
      I've pasted in a chunk of the kstat.h comment at the very bottom
      for later reference...<br>
      &lt;/off-topic&gt;<br>
      <br>
      Because occupancy can be done in units of time, we don't have to
      know buffer and packet *sizes*<br>
      <ul>
        <li>A well-performing queue/buffer system will have an average
          queue occupancy of &lt; 1</li>
        <li>A buffer is of sufficient size if the queue occupancy
          regularly falls to zero repeatedly during the time period.
          Another way of saying the same thing is that&nbsp; the queue
          empties.</li>
        <li>An operable queue management system sends congestion
          notifications if the queue/buffer does not empty</li>
        <li>A good queue management system sends congestion
          notifications in time to keep the queue/buffer emptying
          regularly.</li>
      </ul>
      The parameter here is the time period, and has to do with the
      maximum rate at which packets (work) can arrive. In other words,
      it's a parameter based on the speed of the link.<br>
      <br>
      --dave<br>
      [By the way, there are lots of better queue-heads than I.&nbsp; The
      serious mathies have way better intuition about what you're seeing
      that I do]<br>
      <br>
      <blockquote
cite="mid:CAA93jw4Goh3Y4OrwTJDUcBGN0bj6YEAm++MYuAaqXiS4Er6q+g@mail.gmail.com"
        type="cite">
        <pre wrap="">
</pre>
        <blockquote type="cite">
          <blockquote type="cite">
            <pre wrap="">On Jan 17, 2014, at 3:51 PM, David Collier-Brown <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:davec-b@rogers.com">&lt;davec-b@rogers.com&gt;</a> wrote:

I've been reading through the internet-drafts, and one paragraph struck me
as very illuminating.  This is therefor a sanity-check before I go full-hog
down a particular path...

The comment is from Baker and Fairhurst,
<a moz-do-not-send="true" class="moz-txt-link-freetext" href="https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/">https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/</a> and the
paragraph is [emphases added]

The point of buffering in the network is to absorb data bursts and to
transmit them during the (hopefully) ensuing bursts of silence. This is
essential to permit the transmission of bursty data. Normally small queues
are preferred in network devices, with sufficient queue capacity to absorb
the bursts. The counter-intuitive result is that maintaining normally-small
queues can result in higher throughput as well as lower end-to- end delay.
In summary, queue limits should not reflect the steady state queues we want
to be maintained in the network; instead, they should reflect the size of
bursts that a network device needs to absorb.

All of a sudden we're talking about the kinds of queues I know a little
about (:-))
---

I'm going to suggest that these are queues and associated physical buffers
that do two things:

hold packets that arrive at a bottleneck for a long as it takes to send them
out a slower link that they came in on, and
hold bursts of packets that arrive adjacent to each other until they can be
sent out in a normal spacing, with some small amount of time between them


In an illustration of Dave Taht's, the first looks something like this

-------------------------+
    |X|            |X|      +-------------------
    |X|            |X|         |XXX|    |XXX|
    |X|            |X|      +-------------------
-------------------------+

At the choke-point there is a buffer at least big enough to give the packet
a chance to wheel from line into column (:-)) and start down the smaller
pipe.

The speed at which the acks come back,  the frequency of drops, and any
explicit congestion notifications slows the sender until they don't overload
the skinnier pipe, thus spacing the packets in the fatter pipe out.

Various causes [Leland] can slow or speed the packets in the fat pipe,
making it possible for several to arrive adjacent to each other, followed by
a gap.  The second purpose of a buffer is to hold  these bursts while things
space themselves back out.

They need to be big enough at minimum to do the speed matching,  and at
maximum,  big enough to spread a burst back into a normal progression,
always assuming that acks, drops and explicit congestion notifications are
slowing the sender to the speed of the slowest part of the network.
---

If I'm right about this, we can draw some helpful conclusions

buffer sizes can be set based on measurements:

speed differences, which are pretty static, plus
observed burstyness

drops and ECN can be done to match the slowest speed in the path

The latter suddenly sounds a bit like path MTU discovery, except it's a bit
more dynamic, and varies with both path and what's happening in various
parts of it.

To me, as a capacity/performance nerd, this sounds a lot more familiar and
manageable. My question to you, before I start madly scribbling on the
internet drafts is:

    Am I blowing smoke?

--dave

--
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
<a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:davecb@spamcop.net">davecb@spamcop.net</a>           |                      -- Mark Twain
(416) 223-8968

_______________________________________________

aqm mailing list
<a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:aqm@ietf.org">aqm@ietf.org</a>
<a moz-do-not-send="true" class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/aqm">https://www.ietf.org/mailman/listinfo/aqm</a>



</pre>
          </blockquote>
          <pre wrap="">

--
Dave T&auml;ht

Fixing bufferbloat with cerowrt: <a moz-do-not-send="true" class="moz-txt-link-freetext" href="http://www.teklibre.com/cerowrt/subscribe.html">http://www.teklibre.com/cerowrt/subscribe.html</a>
</pre>
        </blockquote>
        <pre wrap="">

</pre>
      </blockquote>
      <br>
      <pre wrap="">       /*
         * Accumulated time and queue length statistics.
         *
         * Accumulated time statistics are kept as a running sum
         * of "active" time.  Queue length statistics are kept as a
         * running sum of the product of queue length and elapsed time
         * at that length -- i.e., a Riemann sum for queue length
         * integrated against time.  (You can also think of the active time
         * as a Riemann sum, for the boolean function (queue_length &gt; 0)
         * integrated against time, or you can think of it as the
         * Lebesgue measure of the set on which queue_length &gt; 0.)
         *
         *              ^
         *              |                       _________
         *              8                       | i4    |
         *              |                       |       |
         *      Queue   6                       |       |
         *      Length  |       _________       |       |
         *              4       | i2    |_______|       |
         *              |       |           i3          |
         *              2_______|                       |
         *              |    i1                         |
         *              |_______________________________|
         *              Time-&gt;  t1      t2      t3      t4
         * At each change of state (entry or exit from the queue),
         * we add the elapsed time (since the previous state change)
         * to the active time if the queue length was non-zero during
         * that interval; and we add the product of the elapsed time
         * times the queue length to the running length*time sum.
         *
         * This method is generalizable to measuring residency
         * in any defined system: instead of queue lengths, think
         * of "outstanding RPC calls to server X".
         *
         * A large number of I/O subsystems have at least two basic
         * "lists" of transactions they manage: one for transactions
         * that have been accepted for processing but for which processing
         * has yet to begin, and one for transactions which are actively
         * being processed (but not done). For this reason, two cumulative
         * time statistics are defined here: wait (pre-service) time,
         * and run (service) time.
         *
         * All times are 64-bit nanoseconds (hrtime_t), as returned by
         * gethrtime().
         *
         * The units of cumulative busy time are accumulated nanoseconds.
         * The units of cumulative length*time products are elapsed time
         * times queue length.        
         */


The ascii art is an excerpt from /usr/include/sys/kstat.h on Solaris 10</pre>
      <br>
      <br>
      <pre class="moz-signature" cols="72">-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
<a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:davecb@spamcop.net">davecb@spamcop.net</a>           |                      -- Mark Twain
(416) 223-8968
</pre>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
aqm mailing list
<a class="moz-txt-link-abbreviated" href="mailto:aqm@ietf.org">aqm@ietf.org</a>
<a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/aqm">https://www.ietf.org/mailman/listinfo/aqm</a>
</pre>
    </blockquote>
    <br>
    <br>
    <pre class="moz-signature" cols="72">-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
<a class="moz-txt-link-abbreviated" href="mailto:davecb@spamcop.net">davecb@spamcop.net</a>           |                      -- Mark Twain
(416) 223-8968
</pre>
  </body>
</html>

--------------070508030001090201030906--

From bob.briscoe@bt.com  Thu Jan 23 06:51:35 2014
Return-Path: <bob.briscoe@bt.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 259E01A0434; Thu, 23 Jan 2014 06:51:35 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.136
X-Spam-Level: 
X-Spam-Status: No, score=-3.136 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, RP_MATCHES_RCVD=-0.535, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id kU-VzVjD4OF6; Thu, 23 Jan 2014 06:51:32 -0800 (PST)
Received: from hubrelay-rd.bt.com (hubrelay-rd.bt.com [62.239.224.99]) by ietfa.amsl.com (Postfix) with ESMTP id A6CA81A045B; Thu, 23 Jan 2014 06:51:31 -0800 (PST)
Received: from EVMHR71-UKRD.domain1.systemhost.net (10.36.3.109) by EVMHR67-UKRD.bt.com (10.187.101.22) with Microsoft SMTP Server (TLS) id 8.3.297.1; Thu, 23 Jan 2014 14:51:30 +0000
Received: from EPHR02-UKIP.domain1.systemhost.net (147.149.100.81) by EVMHR71-UKRD.domain1.systemhost.net (10.36.3.109) with Microsoft SMTP Server (TLS) id 8.3.279.1; Thu, 23 Jan 2014 14:51:28 +0000
Received: from bagheera.jungle.bt.co.uk (132.146.168.158) by EPHR02-UKIP.domain1.systemhost.net (147.149.100.81) with Microsoft SMTP Server id 14.2.347.0; Thu, 23 Jan 2014 14:51:28 +0000
Received: from BTP075694.jungle.bt.co.uk ([10.109.25.144])	by bagheera.jungle.bt.co.uk (8.13.5/8.12.8) with ESMTP id s0NEpQ4D000875;	Thu, 23 Jan 2014 14:51:26 GMT
Message-ID: <201401231451.s0NEpQ4D000875@bagheera.jungle.bt.co.uk>
X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9
Date: Thu, 23 Jan 2014 14:51:25 +0000
To: Ingemar Johansson S <ingemar.s.johansson@ericsson.com>
From: Bob Briscoe <bob.briscoe@bt.com>
In-Reply-To: <81564C0D7D4D2A4B9A86C8C7404A13DA31EC25B8@ESESSMB205.ericss on.se>
References: <201311042203.rA4M3lo0026458@bagheera.jungle.bt.co.uk> <CA7A7C64CC4ADB458B74477EA99DF6F502347353E1@HE111643.EMEA1.CDS.T-INTERNAL.COM> <81564C0D7D4D2A4B9A86C8C7404A13DA31EC250B@ESESSMB205.ericsson.se> <201401212249.s0LMneWw026346@bagheera.jungle.bt.co.uk> <81564C0D7D4D2A4B9A86C8C7404A13DA31EC25B8@ESESSMB205.ericsson.se>
MIME-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"; format=flowed
Content-Transfer-Encoding: quoted-printable
X-Scanned-By: MIMEDefang 2.56 on 132.146.168.158
Cc: "Ruediger.Geib@telekom.de" <Ruediger.Geib@telekom.de>, "tsvwg@ietf.org" <tsvwg@ietf.org>, "aqm@ietf.org" <aqm@ietf.org>
Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or tunnel protocols?
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 Jan 2014 14:51:35 -0000

Ingemar, inline...

At 06:36 22/01/2014, Ingemar Johansson S wrote:
>Hi
>
>Please find answers inline
>
>/Ingemar
>
> > -----Original Message-----
> > From: Bob Briscoe [mailto:bob.briscoe@bt.com]
> > Sent: den 21 januari 2014 23:50
> > To: Ingemar Johansson S; Ruediger.Geib@telekom.de
> > Cc: tsvwg@ietf.org; aqm@ietf.org
> > Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to=
 L2
> > or tunnel protocols?
> >
> > Ingemar,
> >
> > 1) Thx for the pointer. We should add this as another example where a=
 lower
> > layer marks the IP header (the example we already have is the L3 switch=
 in
> > the Ethernet world).
> >
> > 2) It doesn't talk about propagation of ECN markings during encap &=
 decap,
> > which is one of the things ecn-encap-guidelines aims to give guidelines=
 on.
> > Do you think a doc like the draft we've done=20
> would be useful to help 3GPP in
> > this respect? Do you think a formal liaison=20
> would be useful to point it out?
>L2 in this respect are the PDCP, RLC and MAC=20
>layers (to be honest I am not fully clear what=20
>constitutes L2). The problem is that to given=20
>encap/decap a meaning in this case it becomes=20
>necessary to encode at least an ECN-CE bit into=20
>any of the PDCP or RLC or MAC headers, as PDCP=20
>is ciphered it probably needs to be in the RLC=20
>or MAC headers. Currently there are no standards=20
>efforts to add an ECN-CE bit to any of these headers.

I didn't mean encap of L3 in L2, because that=20
would indeed require major protocol changes. In=20
this case, the L2 node is marking L3 directly (see 1 above).

I meant L3 in L3 (tunnelling), which is prevalent in 3GPP.

> >
> > 3) 3GPP TS 36.300 is ambiguous whether ECN=20
> marking is applied to all packets
> > to "indicate congestion", or whether it is applied with a frequency or
> > probability that depends on an AQM (it doesn't mention AQM, altho
> > obviously it refers to RFC3168 that is based on AQM). Do you know=
 whether
> > it was meant to imply use of AQM?
>Yes, it is very ambiguous, I advocate for a=20
>frequency or probability as it makes most sense,=20
>but this is not specified at all. The ECN=20
>marking can imply the use of AQM but that is=20
>again not specified. In short, much of the=20
>actual implementation is vendor specific.

OK. In general vendor-specific is good, and=20
congestion control can be robust to different behaviours.

Our draft states that algorithm guidance is outside its scope, but...

One of the purposes of the IETF's AQM guidance is=20
to set bounds on how much variety is feasible=20
without losing interworking. There's a message in=20
3GPP TS 36.300 on how the IETF's RFCs could be=20
(mis)interpreted, if we don't make them clear.


Bob


> >
> > [Ruediger, thx for the supportive words too]
> >
> >
> > Bob
> >
> > At 20:29 21/01/2014, Ingemar Johansson S wrote:
> > >Hi
> > >
> > >Please note that ECN over LTE radio access is already standardized in
> > >3GPP TS 36.300 (see 11.6 in
> > >http://www.3gpp.org/ftp/specs/archive/36_series/36.300/36300-c00.zip )
> > >
> > >/Ingemar
> > > > -----Original Message-----
> > > > From: Ruediger.Geib@telekom.de [mailto:Ruediger.Geib@telekom.de]
> > > > Sent: den 21 januari 2014 08:38
> > > > To: bob.briscoe@bt.com
> > > > Cc: tsvwg@ietf.org; aqm@ietf.org
> > > > Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN
> > > > to L2 or tunnel protocols?
> > > >
> > > > Hi Bob,
> > > >
> > > > I support the issue being picked up by IETF. What can be done within
> > > > the bounds of IETF responsibility should be done. If ECN is seeing
> > > > deployment, especially ECN support for IP over VLAN over IP/MPLS may
> > be of interest.
> > > > Further, ECN over LTE radio Access may be relevant (but my expertise
> > > > is too limited to judge details).
> > > >
> > > > Regards,
> > > >
> > > > Ruediger
> > > >
> > > > -----Urspr=FCngliche Nachricht-----
> > > > Von: tsvwg-bounces@ietf.org [mailto:tsvwg-bounces@ietf.org] Im
> > > > Auftrag von Bob Briscoe
> > > > Gesendet: Montag, 4. November 2013 23:04
> > > > An: tsvwg IETF list; AQM IETF list
> > > > Cc: draft-briscoe-tsvwg-ecn-encap-guidelines@tools.ietf.org
> > > > Betreff: [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or
> > > > tunnel protocols?
> > > >
> > > > Folks,
> > > >
> > > > Pls respond if you support this being adopted as a work-group item
> > > > in the IETF transport services w-g (tsvwg). The WG
> > > chairs need visibility of interest.
> > > > Even better, if you're willing to read / comment / review /
> > > > implement
> > > >
> > > > Guidelines for Adding Congestion Notification to Protocols that
> > > > Encapsulate IP
> > > >=20
> <http://tools.ietf.org/html/draft-briscoe-tsvwg-ecn-encap-guidelines> > >=
 >
> > > >
> > > > Abstract
> > > >
> > > >     The purpose of this document is to guide the design of=
 congestion
> > > >     notification in any lower layer or tunnelling protocol that
> > > >     encapsulates IP.  The aim is for explicit congestion signals to
> > > >     propagate consistently from lower=20
> layer protocols into IP.  Then the
> > > >     IP internetwork layer can act as a portability layer to carry
> > > >     congestion notification from non-IP-aware congested nodes up to=
 the
> > > >     transport layer (L4).  Following these guidelines should assure
> > > >     interworking between new lower layer congestion notification
> > > >     mechanisms, whether specified by the=20
> IETF or other standards bodies.
> > > >
> > > >
> > > > [Cross-posting tsvwg & aqm, just in case]
> > > >
> > > >
> > > > Bob Briscoe,
> > > > also for co-authors Pat Thaler and John Kaippallimalil
> > > >
> > > >
> > > >
> > __________________________________________________________
> > > > ______
> > > > Bob Briscoe,                                                  BT
> > > >
> > >
> > >_______________________________________________
> > >aqm mailing list
> > >aqm@ietf.org
> > >https://www.ietf.org/mailman/listinfo/aqm
> >
> > __________________________________________________________
> > ______
> > Bob Briscoe,                                                  BT

________________________________________________________________
Bob Briscoe,                                                  BT=20


From bob.briscoe@bt.com  Thu Jan 23 07:15:29 2014
Return-Path: <bob.briscoe@bt.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3D8C01A002E; Thu, 23 Jan 2014 07:15:29 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.136
X-Spam-Level: 
X-Spam-Status: No, score=-3.136 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, RP_MATCHES_RCVD=-0.535, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iVRx68Nc0iRw; Thu, 23 Jan 2014 07:15:23 -0800 (PST)
Received: from hubrelay-rd.bt.com (hubrelay-rd.bt.com [62.239.224.99]) by ietfa.amsl.com (Postfix) with ESMTP id D66671A0006; Thu, 23 Jan 2014 07:15:22 -0800 (PST)
Received: from EVMHR01-UKBR.domain1.systemhost.net (193.113.108.40) by EVMHR68-UKRD.bt.com (10.187.101.23) with Microsoft SMTP Server (TLS) id 8.3.279.1; Thu, 23 Jan 2014 15:15:21 +0000
Received: from EPHR02-UKIP.domain1.systemhost.net (147.149.100.81) by EVMHR01-UKBR.domain1.systemhost.net (193.113.108.40) with Microsoft SMTP Server (TLS) id 8.3.297.1; Thu, 23 Jan 2014 15:15:19 +0000
Received: from bagheera.jungle.bt.co.uk (132.146.168.158) by EPHR02-UKIP.domain1.systemhost.net (147.149.100.81) with Microsoft SMTP Server id 14.2.347.0; Thu, 23 Jan 2014 15:15:17 +0000
Received: from BTP075694.jungle.bt.co.uk ([10.109.25.144])	by bagheera.jungle.bt.co.uk (8.13.5/8.12.8) with ESMTP id s0NFFF6T000987;	Thu, 23 Jan 2014 15:15:16 GMT
Message-ID: <201401231515.s0NFFF6T000987@bagheera.jungle.bt.co.uk>
X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9
Date: Thu, 23 Jan 2014 15:15:14 +0000
To: Fred Baker <fred@cisco.com>, Gorry Fairhurst <gorry@erg.abdn.ac.uk>
From: Bob Briscoe <bob.briscoe@bt.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
X-Scanned-By: MIMEDefang 2.56 on 132.146.168.158
Cc: AQM IETF list <aqm@ietf.org>, tsvwg IETF list <tsvwg@ietf.org>
Subject: [aqm] How AQM Guidance might be misinterpreted
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 Jan 2014 15:15:29 -0000

Fred, Gorry,

In case you've tuned out of the thread below, it's worth reading the 
(very short) section on ECN in 3GPP TS 36.300 to understand how RFCs 
could be misinterpreted (link in email below).

In this case, it's unclear but they seem to be assuming that there 
could be a special ECN marking behaviour on control signalling. It's 
ambiguous whether the 3GPP spec below intends to mean that ECN 
marking of a stream of packets is either all-on or all-off, rather 
than signalled by frequency/probability of marking.

It's perhaps another example to be brought out in the AQM Guidance, 
probably under:
"4.5. AQM algorithms SHOULD NOT be dependent on specific transport 
protocol behaviours"
which currenty has only one example: "don't assume TCP".


Bob


>Date: Thu, 23 Jan 2014 14:51:25 +0000
>To: Ingemar Johansson S <ingemar.s.johansson@ericsson.com>
>From: Bob Briscoe <bob.briscoe@bt.com>
>Subject: RE: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN 
>to  L2 or tunnel protocols?
>Cc: "Ruediger.Geib@telekom.de" <Ruediger.Geib@telekom.de>, 
>"tsvwg@ietf.org" <tsvwg@ietf.org>, "aqm@ietf.org" <aqm@ietf.org>
>
>Ingemar, inline...
>
>At 06:36 22/01/2014, Ingemar Johansson S wrote:
>>Hi
>>
>>Please find answers inline
>>
>>/Ingemar
>>
>> > -----Original Message-----
>> > From: Bob Briscoe [mailto:bob.briscoe@bt.com]
>> > Sent: den 21 januari 2014 23:50
>> > To: Ingemar Johansson S; Ruediger.Geib@telekom.de
>> > Cc: tsvwg@ietf.org; aqm@ietf.org
>> > Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to L2
>> > or tunnel protocols?
>> >
>> > Ingemar,

[snip]

>> > 3) 3GPP TS 36.300 is ambiguous whether ECN marking is applied to 
>> all packets
>> > to "indicate congestion", or whether it is applied with a frequency or
>> > probability that depends on an AQM (it doesn't mention AQM, altho
>> > obviously it refers to RFC3168 that is based on AQM). Do you know whether
>> > it was meant to imply use of AQM?
>>Yes, it is very ambiguous, I advocate for a frequency or 
>>probability as it makes most sense, but this is not specified at 
>>all. The ECN marking can imply the use of AQM but that is again not 
>>specified. In short, much of the actual implementation is vendor specific.
>
>OK. In general vendor-specific is good, and congestion control can 
>be robust to different behaviours.
>
>Our draft states that algorithm guidance is outside its scope, but...
>
>One of the purposes of the IETF's AQM guidance is to set bounds on 
>how much variety is feasible without losing interworking. There's a 
>message in 3GPP TS 36.300 on how the IETF's RFCs could be 
>(mis)interpreted, if we don't make them clear.
>
>Bob
>
>> > At 20:29 21/01/2014, Ingemar Johansson S wrote:
>> > >Hi
>> > >
>> > >Please note that ECN over LTE radio access is already standardized in
>> > >3GPP TS 36.300 (see 11.6 in
>> > >http://www.3gpp.org/ftp/specs/archive/36_series/36.300/36300-c00.zip )
>> > >
>> > >/Ingemar

________________________________________________________________
Bob Briscoe,                                                  BT 


From fred@cisco.com  Thu Jan 23 07:55:59 2014
Return-Path: <fred@cisco.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9D9671A000F; Thu, 23 Jan 2014 07:55:59 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -110.036
X-Spam-Level: 
X-Spam-Status: No, score=-110.036 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RP_MATCHES_RCVD=-0.535, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5, USER_IN_WHITELIST=-100] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id r2Nsf5GZ9fZR; Thu, 23 Jan 2014 07:55:57 -0800 (PST)
Received: from alln-iport-3.cisco.com (alln-iport-3.cisco.com [173.37.142.90]) by ietfa.amsl.com (Postfix) with ESMTP id 8885F1A000A; Thu, 23 Jan 2014 07:55:57 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=4414; q=dns/txt; s=iport; t=1390492557; x=1391702157; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=9pPxRiCkVyAPSI9JjEDaVCUXawJyXNGFAB4byU40uIY=; b=SBV3UwD8+HGmefP0JBdCSoZNbVHNY8tJVveUD5dRXnMUFx9DK8kp2MYJ ONeEEG/+EWzgptbNURpHfTcbklw+3T5cOonGwBiGbH7MN0pxSrzs+gcl/ EijvibWk7xzeneeeeTlOyTB4rM5/Ghnrw+M4mCNzKnch1alNdfYlmFf+F w=;
X-Files: signature.asc : 195
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: AgwFANA64VKtJXHA/2dsb2JhbABbgww4VrtzgRAWdIIlAQEBAwF5BQcEAgEIEQMBAQEBJwcyFAkIAgQOBQ6HbwgNxXAXjh4RASUaEQcGgx6BFASQPIExhjaBMosjhUODLYFxOQ
X-IronPort-AV: E=Sophos;i="4.95,706,1384300800";  d="asc'?scan'208";a="14991615"
Received: from rcdn-core2-5.cisco.com ([173.37.113.192]) by alln-iport-3.cisco.com with ESMTP; 23 Jan 2014 15:55:56 +0000
Received: from xhc-aln-x01.cisco.com (xhc-aln-x01.cisco.com [173.36.12.75]) by rcdn-core2-5.cisco.com (8.14.5/8.14.5) with ESMTP id s0NFtuYb019016 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 23 Jan 2014 15:55:56 GMT
Received: from xmb-rcd-x09.cisco.com ([169.254.9.230]) by xhc-aln-x01.cisco.com ([173.36.12.75]) with mapi id 14.03.0123.003; Thu, 23 Jan 2014 09:55:55 -0600
From: "Fred Baker (fred)" <fred@cisco.com>
To: Bob Briscoe <bob.briscoe@bt.com>
Thread-Topic: How AQM Guidance might be misinterpreted
Thread-Index: AQHPGFOZ1wwzUzR7P0yCO+zyGWHQEA==
Date: Thu, 23 Jan 2014 15:55:55 +0000
Message-ID: <F659186A-D5E9-4F11-AC13-9D88A692CCBF@cisco.com>
References: <201401231515.s0NFFF6T000987@bagheera.jungle.bt.co.uk>
In-Reply-To: <201401231515.s0NFFF6T000987@bagheera.jungle.bt.co.uk>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
x-originating-ip: [10.19.64.117]
Content-Type: multipart/signed; boundary="Apple-Mail=_C1A8AB3C-49D9-4760-9492-E5961522108C"; protocol="application/pgp-signature"; micalg=pgp-sha1
MIME-Version: 1.0
Cc: Gorry Fairhurst <gorry@erg.abdn.ac.uk>, AQM IETF list <aqm@ietf.org>, tsvwg IETF list <tsvwg@ietf.org>
Subject: Re: [aqm] How AQM Guidance might be misinterpreted
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 Jan 2014 15:55:59 -0000

--Apple-Mail=_C1A8AB3C-49D9-4760-9492-E5961522108C
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii


On Jan 23, 2014, at 7:15 AM, Bob Briscoe <bob.briscoe@bt.com> wrote:

> Fred, Gorry,
>=20
> In case you've tuned out of the thread below, it's worth reading the =
(very short) section on ECN in 3GPP TS 36.300 to understand how RFCs =
could be misinterpreted (link in email below).
>=20
> In this case, it's unclear but they seem to be assuming that there =
could be a special ECN marking behaviour on control signalling. It's =
ambiguous whether the 3GPP spec below intends to mean that ECN marking =
of a stream of packets is either all-on or all-off, rather than =
signalled by frequency/probability of marking.
>=20
> It's perhaps another example to be brought out in the AQM Guidance, =
probably under:
> "4.5. AQM algorithms SHOULD NOT be dependent on specific transport =
protocol behaviours"
> which currenty has only one example: "don't assume TCP".

That seems reasonable. I'm not sure, though, what specific change you're =
suggesting. Is it the addition of an example, or rewording of the =
recommendation? Can you give us a specific proposed change?

> Bob
>=20
>=20
>> Date: Thu, 23 Jan 2014 14:51:25 +0000
>> To: Ingemar Johansson S <ingemar.s.johansson@ericsson.com>
>> From: Bob Briscoe <bob.briscoe@bt.com>
>> Subject: RE: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN =
to  L2 or tunnel protocols?
>> Cc: "Ruediger.Geib@telekom.de" <Ruediger.Geib@telekom.de>, =
"tsvwg@ietf.org" <tsvwg@ietf.org>, "aqm@ietf.org" <aqm@ietf.org>
>>=20
>> Ingemar, inline...
>>=20
>> At 06:36 22/01/2014, Ingemar Johansson S wrote:
>>> Hi
>>>=20
>>> Please find answers inline
>>>=20
>>> /Ingemar
>>>=20
>>> > -----Original Message-----
>>> > From: Bob Briscoe [mailto:bob.briscoe@bt.com]
>>> > Sent: den 21 januari 2014 23:50
>>> > To: Ingemar Johansson S; Ruediger.Geib@telekom.de
>>> > Cc: tsvwg@ietf.org; aqm@ietf.org
>>> > Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding =
ECN to L2
>>> > or tunnel protocols?
>>> >
>>> > Ingemar,
>=20
> [snip]
>=20
>>> > 3) 3GPP TS 36.300 is ambiguous whether ECN marking is applied to =
all packets
>>> > to "indicate congestion", or whether it is applied with a =
frequency or
>>> > probability that depends on an AQM (it doesn't mention AQM, altho
>>> > obviously it refers to RFC3168 that is based on AQM). Do you know =
whether
>>> > it was meant to imply use of AQM?
>>> Yes, it is very ambiguous, I advocate for a frequency or probability =
as it makes most sense, but this is not specified at all. The ECN =
marking can imply the use of AQM but that is again not specified. In =
short, much of the actual implementation is vendor specific.
>>=20
>> OK. In general vendor-specific is good, and congestion control can be =
robust to different behaviours.
>>=20
>> Our draft states that algorithm guidance is outside its scope, but...
>>=20
>> One of the purposes of the IETF's AQM guidance is to set bounds on =
how much variety is feasible without losing interworking. There's a =
message in 3GPP TS 36.300 on how the IETF's RFCs could be =
(mis)interpreted, if we don't make them clear.
>>=20
>> Bob
>>=20
>>> > At 20:29 21/01/2014, Ingemar Johansson S wrote:
>>> > >Hi
>>> > >
>>> > >Please note that ECN over LTE radio access is already =
standardized in
>>> > >3GPP TS 36.300 (see 11.6 in
>>> > =
>http://www.3gpp.org/ftp/specs/archive/36_series/36.300/36300-c00.zip )
>>> > >
>>> > >/Ingemar
>=20
> ________________________________________________________________
> Bob Briscoe,                                                  BT=20

------------------------------------------------------
8 issues in virtual infrastructure
http://dcrocker.net/#fallacies


--Apple-Mail=_C1A8AB3C-49D9-4760-9492-E5961522108C
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment; filename="signature.asc"
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Message signed with OpenPGP using GPGMail

-----BEGIN PGP SIGNATURE-----
Comment: GPGTools - http://gpgtools.org

iD8DBQFS4TuIbjEdbHIsm0MRAnccAKC37EFLAHhHR5y+eZfpzFby0/ZAdQCfQ4Iz
HtdXKSL4vabqoMVCaif4eJc=
=DfC4
-----END PGP SIGNATURE-----

--Apple-Mail=_C1A8AB3C-49D9-4760-9492-E5961522108C--

From davec-b@rogers.com  Thu Jan 23 18:42:31 2014
Return-Path: <davec-b@rogers.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 510221A0396 for <aqm@ietfa.amsl.com>; Thu, 23 Jan 2014 18:42:31 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GMtcyXagB35l for <aqm@ietfa.amsl.com>; Thu, 23 Jan 2014 18:42:22 -0800 (PST)
Received: from nm7-vm7.access.bullet.mail.gq1.yahoo.com (nm7-vm7.access.bullet.mail.gq1.yahoo.com [216.39.63.185]) by ietfa.amsl.com (Postfix) with ESMTP id DF2251A035A for <aqm@ietf.org>; Thu, 23 Jan 2014 18:42:21 -0800 (PST)
Received: from [216.39.60.167] by nm7.access.bullet.mail.gq1.yahoo.com with NNFMP; 24 Jan 2014 02:42:20 -0000
Received: from [67.195.22.118] by tm3.access.bullet.mail.gq1.yahoo.com with NNFMP; 24 Jan 2014 02:42:20 -0000
Received: from [127.0.0.1] by smtp113.sbc.mail.gq1.yahoo.com with NNFMP; 24 Jan 2014 02:42:20 -0000
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rogers.com; s=s1024; t=1390531340; bh=27zt3hMdwKtKqrLfBDGvs5M+LgCosoF3mHvCcX9O/SE=; h=X-Yahoo-Newman-Id:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:X-Rocket-Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:Subject:X-Enigmail-Version:Content-Type; b=fcqI4hKFYHWVXbm8U8AufW9tQ/T3132Y6LyR+JB5cHA950r0IhVTMSV+1iQufKWNNyMgdfMtT8xQxGIPTGOyhyf5zam02esYAmRNEChGU1Y61mb1wbMhlae6kGhXc7MSwwnqtl2dXWSLxfxlnoG4nhMvrmDA/oiAPLWqfWyFeyQ=
X-Yahoo-Newman-Id: 851838.74999.bm@smtp113.sbc.mail.gq1.yahoo.com
X-Yahoo-Newman-Property: ymail-3
X-YMail-OSG: 2VhjbcgVM1le5JdN_msVsrANKO0RSyk4rjOX.c20EsjEzCE _rx9xVsGRFeQjJcHihZ8Jc7lxl9IteD4HwYshPjE05YC3oLnkq.G2TbPtIko 1pfnlPD18ONS8Kh2evc51nw9ZE19Fm5jlxIxnQJf_jQPt5P8j.8.xlXBzd2l hHR5kHneiweBdf6JCu688e6f2ymQf2ZNfV58akWc6ijI3ripc7YnJg682ori 52I2lXfwExh9SIXI1x3L1G_5fRlTYMA15P_mWNCf9n64qwM2kgPA9il31ynK cgT8lc3UYVYCaQnBAKCc_jYe85WRYoZwa1pRlxO4jSjEN3fsoKk.4og_lWUA WYkM5045jBxI9BQVuR3iWUhSJK0OxkCQaQpowhorvHorAiGUf8FQEZ8cN5RN LI6BeEHwkMJh1kxPDuCvtV2Hh_Oc7zlMJc9XLT_XYeG502Sm6GtE_MDre_yu gwbXSh7DNOmwunnTLMyXwbxvnapMO2tzwdy_yDPgXwRw4jPoHkL6hxO.tbD0 Z2MWLrdNmR1LF4YyfVj45TRogBqVy8ggLsBpAZnyPZFIsqb4bb_FKbYigrIq 3nSSBytqpCVyyhybDCq6Sg_ur.iCBJ29lcJH.hGZDr85EXDQu2GhQkTpjfDE WjPv0kr01q0gaa2AjHTEf0.yL7Q7XqTn_HAqV_4RbuIWyAMba3RIMsVHqTkC 2ZA--
X-Yahoo-SMTP: sltvjZWswBCRD.ElTuB1l9j6s9wRYPpuyTNWOE5oEg--
X-Rocket-Received: from [192.168.0.10] (davec-b@99.225.150.135 with plain [67.195.22.104]) by smtp113.sbc.mail.gq1.yahoo.com with SMTP; 23 Jan 2014 18:42:20 -0800 PST
Message-ID: <52E1E30A.5010700@rogers.com>
Date: Thu, 23 Jan 2014 22:50:34 -0500
From: David Collier-Brown <davec-b@rogers.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7
MIME-Version: 1.0
To: "aqm@ietf.org" <aqm@ietf.org>
X-Enigmail-Version: 1.6
Content-Type: multipart/alternative; boundary="------------030606040303010503070903"
Subject: [aqm] Comments on draft-ietf-aqm-recommendation-00, message 3 of what turns out to be three
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: davecb@spamcop.net
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 24 Jan 2014 02:42:31 -0000

This is a multi-part message in MIME format.
--------------030606040303010503070903
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

Section, paragraph (page)
=======  =========  ====
4 para 1 (p10)
   The IRTF, in publishing [RFC2309], and the IETF in subsequent 
   discussion, has developed a set of specific recommendations regarding
   the implementation and operational use of AQM procedures.  This
   document updates these to include

As we have a list that is immediately followed by the same list in an
expanded form, I'd suggest a few words to tell the reader what they're
going to see. I'd end the paragraph with "some seven additional
considerations, expanded upon starting in section 4.1

   4, last para (p11)
       However, care should be taken in concluding that one's use  case
falls in that category;


Actually this would be stronger if it was the conclusion of the previous
sentence. I'd suggest the sentence start with "Therefore"

4.1, para 1 (p11) 
   AQM procedures are designed to minimize delay induced in the network by queues that have filled as a result of host behavior.


It's also true hat we're trying to prevent buffer overflow, so I'd
suggest changing "delay" to "delay and buffer exhaustion", to make the
argument for AQM a stronger one.

   4.2, para 5 (p12)
       To be conservative transport must the latter."

I think "assume" may be missing here.


4.5, para 3 
   AQM algorithms SHOULD NOT target or derive implicit assumptions about

   the characteristics desired by specific transports/applications.
   Transports and applications need to respond to the congestion signals
   provided by AQM (i.e. dropping or ECN-marking) in a timely manner
   (within a few RTT at the latest).

This reads like it should be the introductory paragraph of section 4.6, so
I'd move it there.
 

 4.6 para 3, now 4, (p 15)

       A common objective is to deliver data from its source end point to

   its destination in the least possible time.  When speaking of TCP
   performance, the terms "knee" and "cliff" area defined by [Jain94].
   They respectively refer to the minimum congestion window that
   maximises throughput and the maximum congestion window that avoids
   loss.  An application that transmits at the rate determined by this
   window has the effect of maximizing the rate or throughput.  For the
   sender, exceeding the cliff is ineffective, as it (by definition)
   induces loss; operating at a point close to the cliff has a negative
   impact on other traffic and applications, triggering operator
   activities, such as those discussed in [RFC6057].  Operating below
   the knee reduces the throughput, since the sender fails to use
   available network capacity.  As a result, the behavior of any elastic
   transport congestion control algorithm designed to minimise delivery
   time should seek to use an effective window at or above the knee and
   well below the cliff. 

This is not the case where one has a congestion control mechanism, and
it's inconsistent with Jain's later works or a modern understanding
of queues.  The knee is a region, between a gentle increase in latency
with increasing throughput and the cliff, or the "handle of the hockey
stick"
to use modern terminology.   He was using "knee" as the first upturn, and
"cliff" as the completion of the upturn.

Speaking of TCP only, one wants to run up to the point at which one sees
congestion,
then back off, exactly what we try to do. 

I'd therefor say something like

Common objectives of TCP/IP are deliver the maximum amount of data 
without inducing queuing delays when packets sit in a buffer waiting
to be sent on.  It is important to stay below the inflection point
(bend) in the load/delay curve, which is shaped like a hockey-
stick, "_/". Above the bend, congestion exists, delay rises without bound 
and TCP starts dropping packets due to the congestion.

Approaching the bend, delay starts to rise due to congestion, caused
by packets probabalistically arriving at overlapping times. Near-
maximum throughput and low latency are both achieved quite close to the
knee, so it is desirable to send at as rapid a rate as can be done without 
causing congestion.

Choice of an appropriate rate can therfore significantly impact the loss and 
delay experienced not only by a flow, but by other flows that share the same queue.


  This stays away from the possibly fuzzy language and calls out the
packet drop as
something that a queue modeler will recognize as non-standard.

 

Finally, in 6, I'd suggest that a DOS atatck "may probably be disguised
as" rather than "may create"
unresponsive flows that may be indistinguishable from normal
unresponseive ones.
 



End of third day: comments, suggestions and brickbats cordially accepted!

--dave
[If anyone would prefer this as amendments to an editable format, I have
it in a doc file
at https://github.com/davecb/draft-ietf-aqm-recommendation-edits]

--------------030606040303010503070903
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<html>
  <head>
    <meta http-equiv="content-type" content="text/html;
      charset=ISO-8859-1">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <pre>Section, paragraph (page)
=======&nbsp; =========  ====
4 para 1 (p10)
   The IRTF, in publishing [RFC2309], and the IETF in subsequent 
   discussion, has developed a set of specific recommendations regarding
   the implementation and operational use of AQM procedures.  This
   document updates these to include
</pre>
    As we have a list that is immediately followed by the same list in
    an expanded form, I'd suggest a few words to tell the reader what
    they're going to see. I'd end the paragraph with "some seven
    additional
    considerations, expanded upon starting in section 4.1
    <title></title>
    <meta name="GENERATOR" content="LibreOffice 4.1.0.4 (Linux)">
    <style type="text/css">
	<!--
		@page { margin: 0.79in }
		PRE { margin-left: 0.1in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		PRE.western { so-language: en-US }
		PRE.cjk { font-family: "WenQuanYi Zen Hei Sharp", monospace; so-language: zh-CN }
		PRE.ctl { font-family: "Lohit Devanagari", monospace; so-language: hi-IN }
		P { margin-bottom: 0.08in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		P.western { font-family: "Times New Roman", serif; font-size: 12pt; so-language: en-US }
		P.cjk { font-family: "WenQuanYi Zen Hei Sharp"; font-size: 12pt; so-language: zh-CN }
		P.ctl { font-family: "Lohit Devanagari"; font-size: 12pt; so-language: hi-IN }
	-->:"

</style><br>
    <meta http-equiv="CONTENT-TYPE" content="text/html;
      charset=ISO-8859-1">
    <p class="western" style="margin-bottom: 0in">&nbsp;&nbsp; 4, last para (p11)<br>
      &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; However, care should be
      taken in concluding that one's use&nbsp; case falls in that category; </p>
    <title></title>
    <meta name="GENERATOR" content="LibreOffice 4.1.0.4 (Linux)">
    <style type="text/css">
	<!--
		@page { margin: 0.79in }
		PRE { margin-left: 0.1in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		PRE.western { so-language: en-US }
		PRE.cjk { font-family: "WenQuanYi Zen Hei Sharp", monospace; so-language: zh-CN }
		PRE.ctl { font-family: "Lohit Devanagari", monospace; so-language: hi-IN }
		P { margin-bottom: 0.08in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		P.western { font-family: "Times New Roman", serif; font-size: 12pt; so-language: en-US }
		P.cjk { font-family: "WenQuanYi Zen Hei Sharp"; font-size: 12pt; so-language: zh-CN }
		P.ctl { font-family: "Lohit Devanagari"; font-size: 12pt; so-language: hi-IN }
	-->
	</style><br>
    Actually this would be stronger if it was the conclusion of the
    previous sentence. I'd suggest the sentence start with "Therefore"<br>
    <br>
    <pre class="western">4.1, para 1 (p11) 
   AQM procedures are designed to minimize delay induced in the network by queues that have filled as a result of host behavior.</pre>
    <title></title>
    <meta name="GENERATOR" content="LibreOffice 4.1.0.4 (Linux)">
    <style type="text/css">
	<!--
		@page { margin: 0.79in }
		PRE { margin-left: 0.1in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		PRE.western { so-language: en-US }
		PRE.cjk { font-family: "WenQuanYi Zen Hei Sharp", monospace; so-language: zh-CN }
		PRE.ctl { font-family: "Lohit Devanagari", monospace; so-language: hi-IN }
		P { margin-bottom: 0.08in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		P.western { font-family: "Times New Roman", serif; font-size: 12pt; so-language: en-US }
		P.cjk { font-family: "WenQuanYi Zen Hei Sharp"; font-size: 12pt; so-language: zh-CN }
		P.ctl { font-family: "Lohit Devanagari"; font-size: 12pt; so-language: hi-IN }
	-->
	</style><br>
    It's also true hat we're trying to prevent buffer overflow, so I'd
    suggest changing "delay" to "delay and buffer exhaustion", to make
    the argument for AQM a stronger one.<br>
    <br>
    <meta http-equiv="CONTENT-TYPE" content="text/html;
      charset=ISO-8859-1">
    <p class="western" style="margin-bottom: 0in">&nbsp;&nbsp; 4.2, para 5 (p12)<br>
      &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; To be conservative
      transport must the latter."<br>
      <br>
    </p>
    I think "assume" may be missing here.<br>
    <br>
    <br>
    <meta http-equiv="CONTENT-TYPE" content="text/html;
      charset=ISO-8859-1">
    <pre class="western" style="margin-bottom: 0in">4.5, para 3 
   AQM algorithms SHOULD NOT target or derive implicit assumptions about</pre>
    <pre class="western">   the characteristics desired by specific transports/applications.
   Transports and applications need to respond to the congestion signals
   provided by AQM (i.e. dropping or ECN-marking) in a timely manner
   (within a few RTT at the latest).</pre>
    <title></title>
    <meta name="GENERATOR" content="LibreOffice 4.1.0.4 (Linux)">
    <style type="text/css">
	<!--
		@page { margin: 0.79in }
		PRE { margin-left: 0.1in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		PRE.western { so-language: en-US }
		PRE.cjk { font-family: "WenQuanYi Zen Hei Sharp", monospace; so-language: zh-CN }
		PRE.ctl { font-family: "Lohit Devanagari", monospace; so-language: hi-IN }
		P { margin-bottom: 0.08in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		P.western { font-family: "Times New Roman", serif; font-size: 12pt; so-language: en-US }
		P.cjk { font-family: "WenQuanYi Zen Hei Sharp"; font-size: 12pt; so-language: zh-CN }
		P.ctl { font-family: "Lohit Devanagari"; font-size: 12pt; so-language: hi-IN }
	-->
	</style>This reads like it should be the introductory paragraph of
    section 4.6, so<br>
    I'd move it there.<br>
    &nbsp;<br>
    <pre>&nbsp;4.6 para 3, now 4, (p 15)</pre>
    <p class="western" style="margin-bottom: 0in">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; A common
      objective is
      to deliver data from its source end point to</p>
    <pre class="western">   its destination in the least possible time.  When speaking of TCP
   performance, the terms "knee" and "cliff" area defined by [Jain94].
   They respectively refer to the minimum congestion window that
   maximises throughput and the maximum congestion window that avoids
   loss.  An application that transmits at the rate determined by this
   window has the effect of maximizing the rate or throughput.  For the
   sender, exceeding the cliff is ineffective, as it (by definition)
   induces loss; operating at a point close to the cliff has a negative
   impact on other traffic and applications, triggering operator
   activities, such as those discussed in [RFC6057].  Operating below
   the knee reduces the throughput, since the sender fails to use
   available network capacity.  As a result, the behavior of any elastic
   transport congestion control algorithm designed to minimise delivery
   time should seek to use an effective window at or above the knee and
   well below the cliff. </pre>
    <title></title>
    <meta name="GENERATOR" content="LibreOffice 4.1.0.4 (Linux)">
    <style type="text/css">
	<!--
		@page { margin: 0.79in }
		PRE { margin-left: 0.1in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		PRE.western { so-language: en-US }
		PRE.cjk { font-family: "WenQuanYi Zen Hei Sharp", monospace; so-language: zh-CN }
		PRE.ctl { font-family: "Lohit Devanagari", monospace; so-language: hi-IN }
		P { margin-bottom: 0.08in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		P.western { font-family: "Times New Roman", serif; font-size: 12pt; so-language: en-US }
		P.cjk { font-family: "WenQuanYi Zen Hei Sharp"; font-size: 12pt; so-language: zh-CN }
		P.ctl { font-family: "Lohit Devanagari"; font-size: 12pt; so-language: hi-IN }
	-->
	</style>This is not the case where one has a congestion control
    mechanism, and <br>
    it's inconsistent with Jain's later works or a modern understanding<br>
    of queues.&nbsp; The knee is a region, between a gentle increase in
    latency<br>
    with increasing throughput and the cliff, or the "handle of the
    hockey stick"<br>
    to use modern terminology.&nbsp;&nbsp; He was using "knee" as the first
    upturn, and<br>
    "cliff" as the completion of the upturn.<br>
    <br>
    Speaking of TCP only, one wants to run up to the point at which one
    sees congestion,<br>
    then back off, exactly what we try to do.&nbsp; <br>
    <br>
    I'd therefor say something like <br>
    <meta http-equiv="CONTENT-TYPE" content="text/html;
      charset=ISO-8859-1">
    <pre class="western" style="margin-left: 0.49in">
Common objectives of TCP/IP are deliver the maximum amount of data 
without inducing queuing delays when packets sit in a buffer waiting
to be sent on.  It is important to stay below the inflection point
(bend) in the load/delay curve, which is shaped like a hockey-
stick, &#8220;_/&#8221;. Above the bend, congestion exists, delay rises without bound 
and TCP starts dropping packets due to the congestion.

Approaching the bend, delay starts to rise due to congestion, caused
by packets probabalistically arriving at overlapping times. Near-
maximum throughput and low latency are both <font color="#000000"><font face="DejaVu Sans Mono, monospace"><font size="2"><span lang="en-US">achieved</span></font></font></font> quite close to the
knee, so it is desirable to send at as rapid a rate as can be done without 
causing congestion.

Choice of an appropriate rate can therfore significantly impact the loss and 
delay experienced not only by a flow, but by other flows that share the same queue.</pre>
    <br>
    <title></title>
    <meta name="GENERATOR" content="LibreOffice 4.1.0.4 (Linux)">
    <style type="text/css">
	<!--
		@page { margin: 0.79in }
		P { margin-bottom: 0.08in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		P.western { font-family: "Times New Roman", serif; font-size: 12pt; so-language: en-US }
		P.cjk { font-family: "WenQuanYi Zen Hei Sharp"; font-size: 12pt; so-language: zh-CN }
		P.ctl { font-family: "Lohit Devanagari"; font-size: 12pt; so-language: hi-IN }
	-->
	</style>&nbsp;
    <title></title>
    <meta name="GENERATOR" content="LibreOffice 4.1.0.4 (Linux)">
    <style type="text/css">
	<!--
		@page { margin: 0.79in }
		PRE { margin-left: 0.1in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		PRE.western { so-language: en-US }
		PRE.cjk { font-family: "WenQuanYi Zen Hei Sharp", monospace; so-language: zh-CN }
		PRE.ctl { font-family: "Lohit Devanagari", monospace; so-language: hi-IN }
		P { margin-bottom: 0.08in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		P.western { font-family: "Times New Roman", serif; font-size: 12pt; so-language: en-US }
		P.cjk { font-family: "WenQuanYi Zen Hei Sharp"; font-size: 12pt; so-language: zh-CN }
		P.ctl { font-family: "Lohit Devanagari"; font-size: 12pt; so-language: hi-IN }
	-->
	</style>
    This stays away from the possibly fuzzy language and calls out the
    packet drop as<br>
    something that a queue modeler will recognize as non-standard.<br>
    <title></title>
    <meta name="GENERATOR" content="LibreOffice 4.1.0.4 (Linux)">
    <style type="text/css">
	<!--
		@page { margin: 0.79in }
		PRE { margin-left: 0.1in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		PRE.western { so-language: en-US }
		PRE.cjk { font-family: "WenQuanYi Zen Hei Sharp", monospace; so-language: zh-CN }
		PRE.ctl { font-family: "Lohit Devanagari", monospace; so-language: hi-IN }
		P { margin-bottom: 0.08in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		P.western { font-family: "Times New Roman", serif; font-size: 12pt; so-language: en-US }
		P.cjk { font-family: "WenQuanYi Zen Hei Sharp"; font-size: 12pt; so-language: zh-CN }
		P.ctl { font-family: "Lohit Devanagari"; font-size: 12pt; so-language: hi-IN }
	-->
	</style><br>
    &nbsp;
    <meta http-equiv="CONTENT-TYPE" content="text/html;
      charset=ISO-8859-1">
    <p class="western" style="margin-bottom: 0in"> </p>
    <title></title>
    <meta name="GENERATOR" content="LibreOffice 4.1.0.4 (Linux)">
    <style type="text/css">
	<!--
		@page { margin: 0.79in }
		P { margin-bottom: 0.08in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		P.western { font-family: "Times New Roman", serif; font-size: 12pt; so-language: en-US }
		P.cjk { font-family: "WenQuanYi Zen Hei Sharp"; font-size: 12pt; so-language: zh-CN }
		P.ctl { font-family: "Lohit Devanagari"; font-size: 12pt; so-language: hi-IN }
	-->
	</style>Finally, in 6, I'd suggest that a DOS atatck "may probably be
    disguised as" rather than "may create" <br>
    unresponsive flows that may be indistinguishable from normal
    unresponseive ones.<br>
    &nbsp;
    <meta http-equiv="CONTENT-TYPE" content="text/html;
      charset=ISO-8859-1">
    <p class="western" style="margin-bottom: 0in"> </p>
    <title></title>
    <meta name="GENERATOR" content="LibreOffice 4.1.0.4 (Linux)">
    <style type="text/css">
	<!--
		@page { margin: 0.79in }
		P { margin-bottom: 0.08in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		P.western { font-family: "Times New Roman", serif; font-size: 12pt; so-language: en-US }
		P.cjk { font-family: "WenQuanYi Zen Hei Sharp"; font-size: 12pt; so-language: zh-CN }
		P.ctl { font-family: "Lohit Devanagari"; font-size: 12pt; so-language: hi-IN }
	-->
	</style>
    <meta http-equiv="CONTENT-TYPE" content="text/html;
      charset=ISO-8859-1">
    <p class="western" style="margin-bottom: 0in"> </p>
    <title></title>
    <meta name="GENERATOR" content="LibreOffice 4.1.0.4 (Linux)">
    <style type="text/css">
	<!--
		@page { margin: 0.79in }
		P { margin-bottom: 0.08in; direction: ltr; color: #000000; widows: 0; orphans: 0 }
		P.western { font-family: "Times New Roman", serif; font-size: 12pt; so-language: en-US }
		P.cjk { font-family: "WenQuanYi Zen Hei Sharp"; font-size: 12pt; so-language: zh-CN }
		P.ctl { font-family: "Lohit Devanagari"; font-size: 12pt; so-language: hi-IN }
	-->
	</style><br>
    <br>
    End of third day: comments, suggestions and brickbats cordially
    accepted!<br>
    <br>
    --dave<br>
    [If anyone would prefer this as amendments to an editable format, I
    have it in a doc file<br>
    at <a class="moz-txt-link-freetext"
      href="https://github.com/davecb/draft-ietf-aqm-recommendation-edits">https://github.com/davecb/draft-ietf-aqm-recommendation-edits</a>]<br>
  </body>
</html>

--------------030606040303010503070903--

From ingemar.s.johansson@ericsson.com  Thu Jan 30 04:46:42 2014
Return-Path: <ingemar.s.johansson@ericsson.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AB0151A039B; Thu, 30 Jan 2014 04:46:42 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.851
X-Spam-Level: 
X-Spam-Status: No, score=-3.851 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HELO_EQ_SE=0.35, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id otQ5_PUmfu8S; Thu, 30 Jan 2014 04:46:40 -0800 (PST)
Received: from mailgw2.ericsson.se (mailgw2.ericsson.se [193.180.251.37]) by ietfa.amsl.com (Postfix) with ESMTP id 7B1741A03DB; Thu, 30 Jan 2014 04:46:39 -0800 (PST)
X-AuditID: c1b4fb25-b7f038e000005d01-0f-52ea49ab7dad
Received: from ESESSHC007.ericsson.se (Unknown_Domain [153.88.253.124]) by mailgw2.ericsson.se (Symantec Mail Security) with SMTP id FB.A8.23809.BA94AE25; Thu, 30 Jan 2014 13:46:35 +0100 (CET)
Received: from ESESSMB205.ericsson.se ([169.254.5.233]) by ESESSHC007.ericsson.se ([153.88.183.39]) with mapi id 14.02.0387.000; Thu, 30 Jan 2014 13:46:35 +0100
From: Ingemar Johansson S <ingemar.s.johansson@ericsson.com>
To: Bob Briscoe <bob.briscoe@bt.com>
Thread-Topic: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or tunnel protocols?
Thread-Index: AQHPGEqaa95X78tpWEWrA0Il1VF6DZqdPqvg
Date: Thu, 30 Jan 2014 12:46:33 +0000
Message-ID: <81564C0D7D4D2A4B9A86C8C7404A13DA31ED40CF@ESESSMB205.ericsson.se>
References: <201311042203.rA4M3lo0026458@bagheera.jungle.bt.co.uk> <CA7A7C64CC4ADB458B74477EA99DF6F502347353E1@HE111643.EMEA1.CDS.T-INTERNAL.COM> <81564C0D7D4D2A4B9A86C8C7404A13DA31EC250B@ESESSMB205.ericsson.se> <201401212249.s0LMneWw026346@bagheera.jungle.bt.co.uk> <81564C0D7D4D2A4B9A86C8C7404A13DA31EC25B8@ESESSMB205.ericsson.se> <201401231451.s0NEpQ4D000875@bagheera.jungle.bt.co.uk>
In-Reply-To: <201401231451.s0NEpQ4D000875@bagheera.jungle.bt.co.uk>
Accept-Language: sv-SE, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [153.88.183.18]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFrrOLMWRmVeSWpSXmKPExsUyM+Jvje5qz1dBBsfu6Fqs2SdpMX39F0aL D9M4LI69ucvmwOLR9mUyk8eSJT+ZPNpeKgQwR3HZpKTmZJalFunbJXBlPHpzj6Vgh2XFnMcR DYx92l2MnBwSAiYS547eYYSwxSQu3FvP1sXIxSEkcIhRYtqjVmaQhJDAEkaJ70fdQWw2ARuJ lYe+gzWICKhIHNs5A6iGg4NZoFZibn8SSFhYIFniadN1JoiSFIl/a+azQ9hGEncevQZrZRFQ lWiZugNsPK+Ar0Tj/G52iL29zBI7utrBGjgFnCX2XzjBBmIzCshK3P9+jwXEZhYQl7j1ZD4T xNECEkv2nGeGsEUlXj7+xwphK0p8fLWPEaJeT+LG1ClsELa2xLKFr6EWC0qcnPmEZQKj2Cwk Y2chaZmFpGUWkpYFjCyrGNlzEzNz0suNNjECI+fglt+qOxjvnBM5xCjNwaIkzvvhrXOQkEB6 YklqdmpqQWpRfFFpTmrxIUYmDk6pBkb1Q38O2E773+D6V/Pl7qRtPE8yW+a+DPW+K71GoGOL 9XSejjc6L4yUnHxmm3/Ne/PmSYNx1NcSmRPfE/XXRhy6z3LOUkfysXnQyye53Ru3HvCVbt5/ 1PdVrOWvqJf3kyyYfz0UU5e7Ic/gdSI2ISKAfcVL1+iXz9XYN6wXf3/1Siy7vs+iLc1KLMUZ iYZazEXFiQD9B8ylagIAAA==
Cc: "Ruediger.Geib@telekom.de" <Ruediger.Geib@telekom.de>, "tsvwg@ietf.org" <tsvwg@ietf.org>, "aqm@ietf.org" <aqm@ietf.org>
Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to L2 or tunnel protocols?
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 Jan 2014 12:46:42 -0000

Hi

Sorry about the slow response and thanks for the L2/L3 clarification.=20

As regards to the more detailed specifics around ECN, agree, I believe that=
 the AQM WG can be helpful here and give recommendations on how packets sho=
uld be ECN marked.

/Ingemar

> -----Original Message-----
> From: Bob Briscoe [mailto:bob.briscoe@bt.com]
> Sent: den 23 januari 2014 15:51
> To: Ingemar Johansson S
> Cc: Ruediger.Geib@telekom.de; tsvwg@ietf.org; aqm@ietf.org
> Subject: RE: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN to L=
2
> or tunnel protocols?
>=20
> Ingemar, inline...
>=20
> At 06:36 22/01/2014, Ingemar Johansson S wrote:
> >Hi
> >
> >Please find answers inline
> >
> >/Ingemar
> >
> > > -----Original Message-----
> > > From: Bob Briscoe [mailto:bob.briscoe@bt.com]
> > > Sent: den 21 januari 2014 23:50
> > > To: Ingemar Johansson S; Ruediger.Geib@telekom.de
> > > Cc: tsvwg@ietf.org; aqm@ietf.org
> > > Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding ECN
> > > to L2 or tunnel protocols?
> > >
> > > Ingemar,
> > >
> > > 1) Thx for the pointer. We should add this as another example where
> > > a lower layer marks the IP header (the example we already have is
> > > the L3 switch in the Ethernet world).
> > >
> > > 2) It doesn't talk about propagation of ECN markings during encap &
> > > decap, which is one of the things ecn-encap-guidelines aims to give
> guidelines on.
> > > Do you think a doc like the draft we've done
> > would be useful to help 3GPP in
> > > this respect? Do you think a formal liaison
> > would be useful to point it out?
> >L2 in this respect are the PDCP, RLC and MAC layers (to be honest I am
> >not fully clear what constitutes L2). The problem is that to given
> >encap/decap a meaning in this case it becomes necessary to encode at
> >least an ECN-CE bit into any of the PDCP or RLC or MAC headers, as PDCP
> >is ciphered it probably needs to be in the RLC or MAC headers.
> >Currently there are no standards efforts to add an ECN-CE bit to any of
> >these headers.
>=20
> I didn't mean encap of L3 in L2, because that would indeed require major
> protocol changes. In this case, the L2 node is marking L3 directly (see 1
> above).
>=20
> I meant L3 in L3 (tunnelling), which is prevalent in 3GPP.
>=20
> > >
> > > 3) 3GPP TS 36.300 is ambiguous whether ECN
> > marking is applied to all packets
> > > to "indicate congestion", or whether it is applied with a frequency
> > > or probability that depends on an AQM (it doesn't mention AQM, altho
> > > obviously it refers to RFC3168 that is based on AQM). Do you know
> > > whether it was meant to imply use of AQM?
> >Yes, it is very ambiguous, I advocate for a frequency or probability as
> >it makes most sense, but this is not specified at all. The ECN marking
> >can imply the use of AQM but that is again not specified. In short,
> >much of the actual implementation is vendor specific.
>=20
> OK. In general vendor-specific is good, and congestion control can be rob=
ust
> to different behaviours.
>=20
> Our draft states that algorithm guidance is outside its scope, but...
>=20
> One of the purposes of the IETF's AQM guidance is to set bounds on how
> much variety is feasible without losing interworking. There's a message i=
n
> 3GPP TS 36.300 on how the IETF's RFCs could be (mis)interpreted, if we do=
n't
> make them clear.
>=20
>=20
> Bob
>=20
>=20
> > >
> > > [Ruediger, thx for the supportive words too]
> > >
> > >
> > > Bob
> > >
> > > At 20:29 21/01/2014, Ingemar Johansson S wrote:
> > > >Hi
> > > >
> > > >Please note that ECN over LTE radio access is already standardized
> > > >in 3GPP TS 36.300 (see 11.6 in
> > > >http://www.3gpp.org/ftp/specs/archive/36_series/36.300/36300-c00.zi
> > > >p )
> > > >
> > > >/Ingemar
> > > > > -----Original Message-----
> > > > > From: Ruediger.Geib@telekom.de
> [mailto:Ruediger.Geib@telekom.de]
> > > > > Sent: den 21 januari 2014 08:38
> > > > > To: bob.briscoe@bt.com
> > > > > Cc: tsvwg@ietf.org; aqm@ietf.org
> > > > > Subject: Re: [aqm] [tsvwg] Who supports tsvwg adoption of adding
> > > > > ECN to L2 or tunnel protocols?
> > > > >
> > > > > Hi Bob,
> > > > >
> > > > > I support the issue being picked up by IETF. What can be done
> > > > > within the bounds of IETF responsibility should be done. If ECN
> > > > > is seeing deployment, especially ECN support for IP over VLAN
> > > > > over IP/MPLS may
> > > be of interest.
> > > > > Further, ECN over LTE radio Access may be relevant (but my
> > > > > expertise is too limited to judge details).
> > > > >
> > > > > Regards,
> > > > >
> > > > > Ruediger
> > > > >
> > > > > -----Urspr=FCngliche Nachricht-----
> > > > > Von: tsvwg-bounces@ietf.org [mailto:tsvwg-bounces@ietf.org] Im
> > > > > Auftrag von Bob Briscoe
> > > > > Gesendet: Montag, 4. November 2013 23:04
> > > > > An: tsvwg IETF list; AQM IETF list
> > > > > Cc: draft-briscoe-tsvwg-ecn-encap-guidelines@tools.ietf.org
> > > > > Betreff: [tsvwg] Who supports tsvwg adoption of adding ECN to L2
> > > > > or tunnel protocols?
> > > > >
> > > > > Folks,
> > > > >
> > > > > Pls respond if you support this being adopted as a work-group
> > > > > item in the IETF transport services w-g (tsvwg). The WG
> > > > chairs need visibility of interest.
> > > > > Even better, if you're willing to read / comment / review /
> > > > > implement
> > > > >
> > > > > Guidelines for Adding Congestion Notification to Protocols that
> > > > > Encapsulate IP
> > > > >
> > <http://tools.ietf.org/html/draft-briscoe-tsvwg-ecn-encap-guidelines>
> > > > >
> > > > >
> > > > > Abstract
> > > > >
> > > > >     The purpose of this document is to guide the design of conges=
tion
> > > > >     notification in any lower layer or tunnelling protocol that
> > > > >     encapsulates IP.  The aim is for explicit congestion signals =
to
> > > > >     propagate consistently from lower
> > layer protocols into IP.  Then the
> > > > >     IP internetwork layer can act as a portability layer to carry
> > > > >     congestion notification from non-IP-aware congested nodes up =
to
> the
> > > > >     transport layer (L4).  Following these guidelines should assu=
re
> > > > >     interworking between new lower layer congestion notification
> > > > >     mechanisms, whether specified by the
> > IETF or other standards bodies.
> > > > >
> > > > >
> > > > > [Cross-posting tsvwg & aqm, just in case]
> > > > >
> > > > >
> > > > > Bob Briscoe,
> > > > > also for co-authors Pat Thaler and John Kaippallimalil
> > > > >
> > > > >
> > > > >
> > >
> __________________________________________________________
> > > > > ______
> > > > > Bob Briscoe,                                                  BT
> > > > >
> > > >
> > > >_______________________________________________
> > > >aqm mailing list
> > > >aqm@ietf.org
> > > >https://www.ietf.org/mailman/listinfo/aqm
> > >
> > >
> __________________________________________________________
> > > ______
> > > Bob Briscoe,                                                  BT
>=20
> __________________________________________________________
> ______
> Bob Briscoe,                                                  BT


From dave.taht@gmail.com  Thu Jan 30 11:27:22 2014
Return-Path: <dave.taht@gmail.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E44E41A0435 for <aqm@ietfa.amsl.com>; Thu, 30 Jan 2014 11:27:22 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RuYNkmntwaXI for <aqm@ietfa.amsl.com>; Thu, 30 Jan 2014 11:27:19 -0800 (PST)
Received: from mail-qa0-x231.google.com (mail-qa0-x231.google.com [IPv6:2607:f8b0:400d:c00::231]) by ietfa.amsl.com (Postfix) with ESMTP id 9115A1A0296 for <aqm@ietf.org>; Thu, 30 Jan 2014 11:27:19 -0800 (PST)
Received: by mail-qa0-f49.google.com with SMTP id w8so4947376qac.36 for <aqm@ietf.org>; Thu, 30 Jan 2014 11:27:16 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=LQMRPCpBcpbPzPPdxnECo13ve7/MV2ZXH5vEIachu+k=; b=S7BNsC8cOAunrjPuRsqdsv083VzgpEU5jW4JsmGFa4XgWYilEd8euN5ctMHwwJk/jf DozcDhnPwIgjw8hGeq3RYIQlob8ziqTQC6/OMfm2WS+mscwPEx8rz/mLsLPAd/qQ/HHS AlpRZ5+syI/FEsB1d1kJNdDP93s0artkspFOcDZ10eKaW68n5PLy14i+ULx8ZdsxZAbW RaocpNUqltSLEQzktD71ktl1pSgHXkP4Zp7KXMSM6d/GjPvHv4T+0L/Nte4CF72exVsB zWaES/bEjyqr5TDk1kzWdMucXlLUSyaCGv2x+2GGo2QVme76OihLnl9+MUpVTLziSfPh f/jg==
MIME-Version: 1.0
X-Received: by 10.140.98.135 with SMTP id o7mr22886053qge.102.1391110036098; Thu, 30 Jan 2014 11:27:16 -0800 (PST)
Received: by 10.224.42.70 with HTTP; Thu, 30 Jan 2014 11:27:16 -0800 (PST)
In-Reply-To: <1388773064.857924560@apps.rackspace.com>
References: <201312151857.rBFIuuea043478@gateway0.ipv6.occnc.com> <533BE7A9-7804-4A74-BBFB-C75CCE212434@cisco.com> <1388773064.857924560@apps.rackspace.com>
Date: Thu, 30 Jan 2014 11:27:16 -0800
Message-ID: <CAA93jw5Ly4EwcZG3WW8wMio_M6ZCHYaNVTtabFjoTzXJBhj4vA@mail.gmail.com>
From: Dave Taht <dave.taht@gmail.com>
To: David Reed <dpreed@reed.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: "<curtis@ipv6.occnc.com>" <curtis@ipv6.occnc.com>, "aqm@ietf.org" <aqm@ietf.org>, "Fred Baker \(fred\)" <fred@cisco.com>, bloat <bloat@lists.bufferbloat.net>
Subject: Re: [aqm] [e2e] What is a good burst? -- AQM evaluation guidelines
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 Jan 2014 19:27:23 -0000

On Fri, Jan 3, 2014 at 10:17 AM,  <dpreed@reed.com> wrote:
> End-to-end queueing delay (aggregate of delays in all queues except for t=
he
> queues in the endpoints themselves) should typically never (never means
> <99.9% of any hour-long period)  exceed 200 msec. in the worst case, and =
if
> at all possible never exceed 100 msec.

Latency requirements are a budget that can get spent at multiple levels in
a stack. If you were to say that total latency/propagation delay
should not exceed lightspeed requirements + a few ms, I'd agree with
you harder than I already do.

http://gettys.wordpress.com/2013/07/10/low-latency-requires-smart-queuing-t=
raditional-aqm-is-not-enough/

I also generally prefer predictable latency (low jitter), not just from a h=
uman
factors perspective but as a means of doing congestion avoidance
detection better.

> in networks capable of carrying
> more than 1 Mbit/sec to and from endpoints (I would call that high-bitrat=
e
> nets, the stage up from "dialup" networks).

Here are some recent tests of fq_codel on comcast ipv6 over a coast to
coast path (california to maine. Only *65ms* inherent delay - quite
good)

http://snapon.lab.bufferbloat.net/~cero2/jimreisert/results.html

Utilization ~100%, total induced latency under 10ms (some of this
coming from the rate limited upstream and HZ=3D250 cpu scheduler delay
for the rate limiter, not the core algorithms)

the results with pie or codel alone are considerably worse, (40-60ms
induced delay) but not bad compared to the alternative of no
aqm/packet scheduling as shown in the 2nd and 3rd graph.

I'm happy. There seems to be not a lot of slop left to fix in
fq_codel, and amdahls law takes queuing delay out of the latency
equation for applications...

well, aside from getting the CMTSes/dslams/modems fixed... :(

I would additionally distinguish in the range between 100mbit networks
and 1g+ networks, (as well as wireless/wifi/cable) where bursty
technologies are seemingly needed (so long as bursts are below human
perceptible latency factors)

I have a test of an intel nuc with linux 3.11 and pfifo_fast (the
default qdisc), on a gbit lan vs fq_codel, and with fq_codel and
tso/gso/ufo offloading off, lying around somewhere. with pfifo_fast
you end up with one stream using more throughput than the others and
about 8ms latency, fq_codel, 2.2, and without tso/gso you can't
saturate the medium.

I have some hopes the new fq scheduler and some tso fixes make the
results better, I think 2ms of induced queue latency on a gig lan is a lot.

> There are two reasons for this:
> 1) round-trip "RPC" response time for interactive applications > 100 msec=
.
> become unreasonable.

Add in the whole list of human factors issues noted in the url above.

> 2) flow control at the source that stanches the entry of data into the
> network (which can be either switching media codecs or just pushing back =
on
> the application rate - whether it is driven by the receiver or the sender=
,
> both of which are common) must respond quickly, lest more packets be dump=
ed
> into the network that sustain congestion.

A big problem is predictability. recently there has been something of a pus=
h
to get retries and retransmits on wifi to "give up" at 50ms induced latency=
,
instead of 250ms (which some vendors try to do)

We have an inherent quantum problem of 1-4ms per txop in present
day wifi that seems to make 50ms a barely achievable outer limit in
the presence of multiple stations.

> Fairness is a different axis, but I strongly suggest that there are other
> ways to achieve approximate fairness of any desired type without building=
 up
> queues in routers.  It's perfectly reasonable to remember (in all the mem=
ory
> that *would otherwise have caused trouble by holding packets rather than
> discarding them*) the source/dest information and sizes of recently
> processed (forwarded or discarded) packets.  This information takes less
> space than the packets themselves, of course!  It can even be further
> compressed by "coding or hashing" techniques.  Such live data about *rece=
nt
> behavior* is all you need for fairness in balancing signaling back to the
> source.

I concur. Long on my todo list for *codel has been gaining the ability to
toss drop/mark/current bandwidth information on packets up to userspace,
where it could be used to make more intelligent routing decisions, and/or
feed more information back into the senders. I don't think "source quench"
is going to work, tho...

I am encouraged by recent work in openflow in this area.

> If all of the brainpower on this list cannot take that previous paragraph
> and expand it to implement the solution I am talking about, I would be ha=
ppy
> (at my consulting rates, which are quite high) to write the code for you.
> But I have a day job that involves low-level scheduling and queueing work=
 in
> a different domain of application.
>
>
>
> Can we please get rid of the nonsense that implies that the only informat=
ion
> one can have at a router/switch is the set of packets that are clogging i=
ts
> outbound queues?  Study some computer algorithms that provide memory of
> recent history....

Multiple devices are still constrained by memory and cpu and the
needs for low latency insides those devices. At 10Gig you have ns
to make decisions in. On 802.11ac, you have us.

Thus we end up with dedicated hardware and software doing these
jobs that have issues doing stuff in-band, or even out of band.

that said, softer routers are doing hundreds of gig, 802.11ac devices
contain a lot of hardware assist and a dedicated cpu (but closed
firmware), and
the future looks bright for smarter hardware.

btw:

we have (at least temporarily) hit a performance wall on the hardware we
use in cerowrt - not in the aqm algorithms, but in the software rate
limiter, which peaks out currently at about 60mbits. Something faster
than the
current htb algo is needed... (suggestions?) (or a switch to faster
hardware than the 7 year old chipset cero uses)

> and please, please, please stop insisting that
> intra-network queues should build up for any reason whatsoever other than
> instantaneous transient burstiness of convergent traffic.  They should
> persist as briefly as possible, and not be sustained for some kind of
> "optimum" throughput that can be gained by reframing the problem.

well I outlined that bursts are needed in some technologies to keep them
operating at good throughput. Thus, BQL for linux ethernet, and some
similar techniques under consideration for wifi.

but largely agree. Anybody want to aim for 5ms queue delay on
intercontinental links?

>
>
>
>
>
>
>
>
>
> On Thursday, January 2, 2014 1:31am, "Fred Baker (fred)" <fred@cisco.com>
> said:
>
>>
>> On Dec 15, 2013, at 10:56 AM, Curtis Villamizar <curtis@ipv6.occnc.com>
>> wrote:
>>
>> > So briefly, my answer is: as a WG, I don't think we want to go there.
>> > If we do go there at all, then we should define "good AQM" in terms of
>> > acheving a "good" tradeoff between fairness, bulk transfer goodput,
>> > and bounded delay. IMHO sometimes vague is better.
>>
>> As you may have worked out from my previous comments in these threads, I
>> agree
>> with you. I don't think this can be nailed down in a universal sense. Wh=
at
>> can be
>> described is the result in the network, in that delays build up that
>> persist, as
>> opposed to coming and going, and as a result applications don't work as
>> well as
>> they might - and at that point, it is appropriate for the network to
>> inform the
>> transport.
>>
>
>
> _______________________________________________
> aqm mailing list
> aqm@ietf.org
> https://www.ietf.org/mailman/listinfo/aqm
>



--=20
Dave T=E4ht

Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.=
html

From georgehanes@hushmail.com  Thu Jan 30 14:04:34 2014
Return-Path: <georgehanes@hushmail.com>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9405E1A04DB for <aqm@ietfa.amsl.com>; Thu, 30 Jan 2014 14:04:34 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.323
X-Spam-Level: 
X-Spam-Status: No, score=-2.323 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, FREEMAIL_FROM=0.001, RP_MATCHES_RCVD=-0.535, SPF_HELO_NEUTRAL=0.112, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Oj_I_3Oi0fo3 for <aqm@ietfa.amsl.com>; Thu, 30 Jan 2014 14:04:33 -0800 (PST)
Received: from smtp3.hushmail.com (smtp3a.hushmail.com [65.39.178.201]) by ietfa.amsl.com (Postfix) with ESMTP id 39D6C1A04DA for <aqm@ietf.org>; Thu, 30 Jan 2014 14:04:33 -0800 (PST)
Received: from smtp3.hushmail.com (smtp3a.hushmail.com [65.39.178.201]) by smtp3.hushmail.com (Postfix) with SMTP id 02B54E03B5 for <aqm@ietf.org>; Thu, 30 Jan 2014 22:04:23 +0000 (UTC)
Received: from smtp.hushmail.com (w8.hushmail.com [65.39.178.52]) by smtp3.hushmail.com (Postfix) with ESMTP for <aqm@ietf.org>; Thu, 30 Jan 2014 22:04:22 +0000 (UTC)
Received: by smtp.hushmail.com (Postfix, from userid 99) id E31396018A; Thu, 30 Jan 2014 22:04:22 +0000 (UTC)
MIME-Version: 1.0
Date: Thu, 30 Jan 2014 17:04:22 -0500
To: aqm@ietf.org
From: georgehanes@hushmail.com
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset="UTF-8"
Message-Id: <20140130220422.E31396018A@smtp.hushmail.com>
Subject: [aqm] Be cautious of this computer science conference
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 Jan 2014 22:11:35 -0000

Be cautious of this computer science conference

If you have any thought of attending the worldâ€™s biggest 
f-a-k-e conference in computer science 
http://www.world-academy-of-science.org  you should visit 
any websites below

https://sites.google.com/site/worlddump1 
or
https://sites.google.com/site/dumpconf 
https://sites.google.com/site/moneycomp1
https://sites.google.com/site/worlddump4

The organizer of this conference is H-amid A-rabnia  
http://www.cs.uga.edu/~hra  a professor from University 
of Georgia, Athens, US.  He already earned millions of 
dollars from the registration fee. He recently started 
a new conference CSCI due to his hunger for money 
http://www.americancse.org 

He did not reveal the reviews and reviewers' information 
for all the papers he received, despite repeated requests 
and challenges. The reason for his failure is there are 
no reviews and reviewers and he just cheated the research 
community for more than a decade by announcing that each 
draft paper is reviewed by two experts. We challenge him 
to publish these details at the conference website. 
Where are your experts? Where are your reviews? 

Soon he comes up with a story announcing that he lost all 
the information having reviews and reviewers because of 
computer crash or theft.

DBLP stopped indexing these conferences since 2011 and 
displayed an explicit message; 
"The DBLP Advisory Board decided to discontinue indexing 
of this conference series". Visit 
http://www.informatik.uni-trier.de/~ley/db/conf/biocomp/index.html 
as a sample.

He was forced to remove his name, the university of Georgia 
name, and university of Georgia email address from the 
conferenceâ€™s contact page because the University has 
banned him from doing that. Do not spoil your resume by 
publishing in this conference.

Apologies for posting to multiple mailing lists. Spreading 
the news is the only way to stop this conference from 
harming innocent researchers.

Respectfully,

Many researchers cheated by these conferences


From internet-drafts@ietf.org  Thu Jan 30 14:20:36 2014
Return-Path: <internet-drafts@ietf.org>
X-Original-To: aqm@ietfa.amsl.com
Delivered-To: aqm@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BD7201A0504; Thu, 30 Jan 2014 14:20:36 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id lc1G8FBPj9Sx; Thu, 30 Jan 2014 14:20:35 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 57F931A04DA; Thu, 30 Jan 2014 14:20:35 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
From: internet-drafts@ietf.org
To: i-d-announce@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 4.90.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20140130222035.13626.23437.idtracker@ietfa.amsl.com>
Date: Thu, 30 Jan 2014 14:20:35 -0800
Cc: aqm@ietf.org
Subject: [aqm] I-D Action: draft-ietf-aqm-recommendation-01.txt
X-BeenThere: aqm@ietf.org
X-Mailman-Version: 2.1.15
List-Id: "Discussion list for active queue management and flow isolation." <aqm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aqm>, <mailto:aqm-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aqm/>
List-Post: <mailto:aqm@ietf.org>
List-Help: <mailto:aqm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aqm>, <mailto:aqm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 Jan 2014 22:20:37 -0000

A New Internet-Draft is available from the on-line Internet-Drafts director=
ies.
 This draft is a work item of the Active Queue Management and Packet Schedu=
ling Working Group of the IETF.

        Title           : IETF Recommendations Regarding Active Queue Manag=
ement
        Authors         : Fred Baker
                          Godred Fairhurst
	Filename        : draft-ietf-aqm-recommendation-01.txt
	Pages           : 22
	Date            : 2014-01-30

Abstract:
   This memo presents recommendations to the Internet community
   concerning measures to improve and preserve Internet performance.  It
   presents a strong recommendation for testing, standardization, and
   widespread deployment of active queue management (AQM) in network
   devices, to improve the performance of today's Internet.  It also
   urges a concerted effort of research, measurement, and ultimate
   deployment of AQM mechanisms to protect the Internet from flows that
   are not sufficiently responsive to congestion notification.

   The note largely repeats the recommendations of RFC 2309, updated
   after fifteen years of experience and new research.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-aqm-recommendation/

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-aqm-recommendation-01

A diff from the previous version is available at:
http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-aqm-recommendation-01


Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at tools.ietf.org.

Internet-Drafts are also available by anonymous FTP at:
ftp://ftp.ietf.org/internet-drafts/

