
From richard.alimi@gmail.com  Wed Feb  1 00:04:26 2012
Return-Path: <richard.alimi@gmail.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D74C111E80C1 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 00:04:26 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.677
X-Spam-Level: 
X-Spam-Status: No, score=-2.677 tagged_above=-999 required=5 tests=[AWL=-0.300, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, J_CHICKENPOX_44=0.6, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id aNuw2-iT2dKe for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 00:04:26 -0800 (PST)
Received: from mail-iy0-f172.google.com (mail-iy0-f172.google.com [209.85.210.172]) by ietfa.amsl.com (Postfix) with ESMTP id D9C9911E80BE for <alto@ietf.org>; Wed,  1 Feb 2012 00:04:25 -0800 (PST)
Received: by iagf6 with SMTP id f6so1419389iag.31 for <alto@ietf.org>; Wed, 01 Feb 2012 00:04:25 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:cc:content-type :content-transfer-encoding; bh=zX83COR973q0nbGYdEHQ64bL5QxJyvrk9OS4eWZM87I=; b=hpCLp6lnKz3BREjxhxP7MOdjVsQwv2jeTs0lvj0w6us7xwsJqFuGUGKr5GB0sV925Y wZjiWwhOO5/Q/xccK6j4Kv8G9VWD4I1Mlc8Y/Kc72xZ8yBtjG4ohyG3Z3Oq67rYxzZz3 vQiKaC0s88bj/JxFH9PIhfSiaMF0g6eAbwVAk=
Received: by 10.42.168.197 with SMTP id x5mr20499547icy.6.1328083465388; Wed, 01 Feb 2012 00:04:25 -0800 (PST)
MIME-Version: 1.0
Sender: richard.alimi@gmail.com
Received: by 10.231.69.7 with HTTP; Wed, 1 Feb 2012 00:04:04 -0800 (PST)
In-Reply-To: <A35E0DC7-C133-480C-9967-EC62CAE9BCA0@niven-jenkins.co.uk>
References: <A35E0DC7-C133-480C-9967-EC62CAE9BCA0@niven-jenkins.co.uk>
From: Richard Alimi <rich@velvetsea.net>
Date: Wed, 1 Feb 2012 00:04:04 -0800
X-Google-Sender-Auth: 7V1g1rT-d3kkKK5w2qzYyNJLSQg
Message-ID: <CA+cvDaZGBS+WM9W=U_rf9qK6w+Ui=vrYTAp9UxKFb4JM+-=MTw@mail.gmail.com>
To: Ben Niven-Jenkins <ben@niven-jenkins.co.uk>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 08:04:26 -0000

I guess I'm a bit late to the party, so I can keep this simple and say
how the draft was currently intended:
(1) The choice of the term "endpoint" (those thingy's that are
contained in PIDs) was not an accident :)
(2) As such, each entry of the Cost Map is the *end-to-end* cost from
the source to the destination (I think this is what Ben meant by
"mesh").  That is, the ALTO Server has already added up the costs on
the physical links along the path for you.
(3) If there is an entry missing from the Cost Map, it means the ALTO
Server doesn't know the cost from that source to destination (or isn't
willing to tell you). I guess you could generate a path, but the point
about whether a packet can actually traverse that path is a good one
-- ALTO doesn't make any claims on routing capabilities offered by
endpoints.

Given that particular meaning, you could imagine doing overlay routing
on that (e.g., in a P2P swarm). In that sense, each endpoint is a node
on a graph, and you could imagine doing shortest path on that.
However, that's quite different than doing shortest path on the actual
network topology.  This was the reason for my comments a few times (I
guess not on the list, it was probably at the mic during a WG session)
that if we were to ever expose raw network topology via ALTO, then we
need to explicitly denote that it would be a different interpretation
of the Cost Map than we have today (an adjacency matrix seems like a
reasonable model here).

Along those lines, I'm a little bit hesitant of the proposal to add a
"transit" attribute to a PID, since that seems to conflate the two
possible interpretations of the cost map (end-to-end costs vs.
adjacency matrix).  If we wanted to convey the actual graph, how about
have an attribute for the entire cost map that indicates explicitly
that it is to be interpreted as an adjacency matrix (and obviously,
ordinal costs would not make sense there).

Given this discussion and the confusion it caused, we'll have to go
back and ensure this is explicitly stated in the draft.

Thoughts?
Rich

On Mon, Jan 30, 2012 at 9:03 AM, Ben Niven-Jenkins
<ben@niven-jenkins.co.uk> wrote:
> Colleagues,
>
> In the current specification of ALTO, are costs always End to End?
>
> What I mean by that is when looking at ALTO cost maps is it possible to s=
afely assume that of there is a cost between PIDX & PID Y and a cost betwee=
n PIDY & PIDZ then the cost between PIDX & PIDZ can be calculated as cost(P=
IDX,PIDY)+cost(PIDY,PIDZ)? [If this assumption does hold, it is obviously n=
ot applicable to ordinal cost types].
>
> I suspect the answer is no, but I wanted to check what the definitive ans=
wer is.
>
>
> For example if a cost map contains:
>
> =A0 "map" : {
> =A0 =A0 "PID1": { "PID2": 1 },
> =A0 =A0 "PID2": { "PID1": 1, "PID3": 2 },
> =A0 =A0 "PID3": { "PID2": 2 }
>
> Can one assume that the cost between PID1 & PID3 is 3 (PID1->PID2 + PID2-=
>PID3)?
>
> How about if the cost map contains:
>
> =A0 "map" : {
> =A0 =A0 "PID1": { "PID2": 1, "PID3": 3 },
> =A0 =A0 "PID2": { "PID1": 1, "PID3": 2 },
> =A0 =A0 "PID3": { "PID2": 2, "PID1": 3 }
>
> Can one assume PID2 is on a path between PID1 & PID3?
>
> Thanks
> Ben
>
> _______________________________________________
> alto mailing list
> alto@ietf.org
> https://www.ietf.org/mailman/listinfo/alto

From jmedved@cisco.com  Wed Feb  1 00:08:09 2012
Return-Path: <jmedved@cisco.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AB79311E80C1 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 00:08:09 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.299
X-Spam-Level: 
X-Spam-Status: No, score=-6.299 tagged_above=-999 required=5 tests=[AWL=0.300,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id K8pY9gcDv9Qr for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 00:08:08 -0800 (PST)
Received: from mtv-iport-1.cisco.com (mtv-iport-1.cisco.com [173.36.130.12]) by ietfa.amsl.com (Postfix) with ESMTP id CC72211E80BE for <alto@ietf.org>; Wed,  1 Feb 2012 00:08:08 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=jmedved@cisco.com; l=2128; q=dns/txt; s=iport; t=1328083688; x=1329293288; h=subject:mime-version:from:in-reply-to:date:cc: content-transfer-encoding:message-id:references:to; bh=BAXrOYVPgRiSS5tAkg8hpu44sVcqqFyVhzBBflAl9rU=; b=m9oU+iX/ffF2JqknGqgyy1xUlBXkDK0gbyqPBa7SU2zyfiMRu5fWK+8r CRhFHrJ2/D1E+BMExiHS9Mhcmgm/64hVwT1cTzhSYBvICla7vUcUSR7vs w1GQ0U29q4e3gPjrSuZXyX7ofniQlJlEo4w1UGSyx/bRB0bdXG/a7eXuj E=;
X-IronPort-AV: E=Sophos;i="4.71,601,1320624000"; d="scan'208";a="26445722"
Received: from mtv-core-2.cisco.com ([171.68.58.7]) by mtv-iport-1.cisco.com with ESMTP; 01 Feb 2012 08:08:08 +0000
Received: from sjc-vpn6-1553.cisco.com (sjc-vpn6-1553.cisco.com [10.21.126.17]) by mtv-core-2.cisco.com (8.14.3/8.14.3) with ESMTP id q11888ub030040; Wed, 1 Feb 2012 08:08:08 GMT
Mime-Version: 1.0 (Apple Message framework v1251.1)
Content-Type: text/plain; charset=windows-1252
From: Jan Medved <jmedved@cisco.com>
In-Reply-To: <E37CB818-879F-4087-BBD2-C9A9A13F2846@cisco.com>
Date: Wed, 1 Feb 2012 00:08:07 -0800
Content-Transfer-Encoding: quoted-printable
Message-Id: <541EF782-09B8-4CF0-934F-F697816CFFF7@cisco.com>
References: <A35E0DC7-C133-480C-9967-EC62CAE9BCA0@niven-jenkins.co.uk> <4F272268.4070804@bell-labs.com> <6A051A1C-3C3E-43C7-888C-AB1E98CB2A86@niven-jenkins.co.uk> <489EF99B-832D-4B52-B8ED-5F1612F8BFE2@cisco.com> <4F2803E7.1090000@bell-labs.com> <E37CB818-879F-4087-BBD2-C9A9A13F2846@cisco.com>
To: stefano previdi <sprevidi@cisco.com>
X-Mailer: Apple Mail (2.1251.1)
Cc: alto <alto@ietf.org>
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 08:08:09 -0000

On Jan 31, 2012, at 7:29 AM, stefano previdi wrote:

>=20
> On Jan 31, 2012, at 4:08 PM, Vijay K. Gurbani wrote:
>=20
>> On 01/31/2012 05:14 AM, stefano previdi wrote:
>> ben> This might be a case where we'd need a PID property to =
distinguish
>> ben> "transit" PIDs from "End Host" PIDs.
>> ben>
>> stefano> sounds pretty reasonable to me.
>>=20
>> Ben: I suspect that your question had a foundation of some sort
>> and was not simply a fishing expedition ;-)
>>=20
>> Let me ask you this: is the PID property that you would like to put
>> in useful for CDN-usage in ALTO?
>=20
>=20
> there are a variety of cases where this would be useful. When=20
> you distribute a link-state topology you know you're talking=20
> about routers and links and semantic is well-known. Same when=20
> you advertise BGP routes or VPNv4 routes or whatever... the=20
> protocol has the semantic you need to consider.
>=20
> In ALTO you have the necessary bits to tell if cost is based on=20
> routing or air miles or others but you don't really know what a=20
> PID is (other than a bunch of addresses).
>=20
Right, but the costs (and cost types) are associated with cost maps, not =
network maps. I still think a "transit" PID does not make sense (while =
thinking that PID attributes are a good thing).=20

A while back we did model link-state topology in ALTO, and the only =
sensible way we could figure out was to assign a PID to each router =
interface and to model a router as a set of PID where costs between the =
"interface" PIDs were 0. Links were modeled in the the Cost Map, where =
the link cost mapped directly into the cost between two "interface" =
PIDs. It did work, but then BGP-LS came along, which was a much cleaner =
way to convert topology.=20

If we want to convey topology to apps via ALTO, we have to come up with =
a convention, so that the receiving app can properly interpret it. =
Stefano, Rich W, Jon and I proposed such convention a couple of IETFs =
ago, but there was lot of resistance to it at that time. Maybe the time =
is right for it this time=85


/Jan



From sprevidi@cisco.com  Wed Feb  1 00:57:13 2012
Return-Path: <sprevidi@cisco.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0C10B21F8735 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 00:57:13 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.199
X-Spam-Level: 
X-Spam-Status: No, score=-102.199 tagged_above=-999 required=5 tests=[AWL=-0.200, BAYES_00=-2.599, J_CHICKENPOX_44=0.6, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1Gm-z6BUMqEJ for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 00:57:12 -0800 (PST)
Received: from av-tac-bru.cisco.com (weird-brew.cisco.com [144.254.15.118]) by ietfa.amsl.com (Postfix) with ESMTP id 26CA821F8734 for <alto@ietf.org>; Wed,  1 Feb 2012 00:57:11 -0800 (PST)
X-TACSUNS: Virus Scanned
Received: from stew-brew.cisco.com (localhost.cisco.com [127.0.0.1]) by av-tac-bru.cisco.com (8.13.8+Sun/8.13.8) with ESMTP id q118v92c011347 for <alto@ietf.org>; Wed, 1 Feb 2012 09:57:09 +0100 (CET)
Received: from dhcp-10-61-98-38.cisco.com (dhcp-10-61-98-38.cisco.com [10.61.98.38]) by stew-brew.cisco.com (8.13.8+Sun/8.13.8) with ESMTP id q118v7AM028536; Wed, 1 Feb 2012 09:57:08 +0100 (CET)
Mime-Version: 1.0 (Apple Message framework v1251.1)
Content-Type: text/plain; charset=iso-8859-1
From: stefano previdi <sprevidi@cisco.com>
In-Reply-To: <CA+cvDaZGBS+WM9W=U_rf9qK6w+Ui=vrYTAp9UxKFb4JM+-=MTw@mail.gmail.com>
Date: Wed, 1 Feb 2012 09:57:07 +0100
Content-Transfer-Encoding: quoted-printable
Message-Id: <1D8ECFB9-C14F-4661-9B68-F2DEE9FEA530@cisco.com>
References: <A35E0DC7-C133-480C-9967-EC62CAE9BCA0@niven-jenkins.co.uk> <CA+cvDaZGBS+WM9W=U_rf9qK6w+Ui=vrYTAp9UxKFb4JM+-=MTw@mail.gmail.com>
To: Richard Alimi <rich@velvetsea.net>
X-Mailer: Apple Mail (2.1251.1)
Cc: alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 08:57:13 -0000

Hi Rich,

On Feb 1, 2012, at 9:04 AM, Richard Alimi wrote:
> I guess I'm a bit late to the party, so I can keep this simple and say
> how the draft was currently intended:
> (1) The choice of the term "endpoint" (those thingy's that are
> contained in PIDs) was not an accident :)
> (2) As such, each entry of the Cost Map is the *end-to-end* cost from
> the source to the destination (I think this is what Ben meant by
> "mesh").  That is, the ALTO Server has already added up the costs on
> the physical links along the path for you.


do you intend that a cost map is supposed to be a full mesh of leaf=20
nodes (as per analogy to SPF algorithm) ? IOW: you're not expected to=20
transit across PIDs ?


> (3) If there is an entry missing from the Cost Map, it means the ALTO
> Server doesn't know the cost from that source to destination (or isn't
> willing to tell you). I guess you could generate a path, but the point
> about whether a packet can actually traverse that path is a good one
> -- ALTO doesn't make any claims on routing capabilities offered by
> endpoints.
>=20
> Given that particular meaning, you could imagine doing overlay routing
> on that (e.g., in a P2P swarm). In that sense, each endpoint is a node
> on a graph, and you could imagine doing shortest path on that.
> However, that's quite different than doing shortest path on the actual
> network topology.  This was the reason for my comments a few times (I
> guess not on the list, it was probably at the mic during a WG session)
> that if we were to ever expose raw network topology via ALTO, then we
> need to explicitly denote that it would be a different interpretation
> of the Cost Map than we have today (an adjacency matrix seems like a
> reasonable model here).


I believe we have two orthogonal elements:

1. Information that can be leveraged by traditional SPF/Dijktra=20
   algorithm so to compute a topology using same algorithm/paradigms=20
   than the ones used in the network. This can be a cost map, I don't=20
   mind, as long as the semantic of the information is compatible with=20=

   what we want to do with it. In this respect, Ben asked a valid=20
   question.

2. The ability, for an ALTO server, to deliver a partial or aggregated=20=

   or virtual topology that would NOT represent the reality of the=20
   network infrastructure but rather the topology that the application=20=

   needs in order to optimize its business.

Implementors have worked quite a bit on the latter.


> Along those lines, I'm a little bit hesitant of the proposal to add a
> "transit" attribute to a PID, since that seems to conflate the two
> possible interpretations of the cost map (end-to-end costs vs.
> adjacency matrix).  If we wanted to convey the actual graph, how about
> have an attribute for the entire cost map that indicates explicitly
> that it is to be interpreted as an adjacency matrix (and obviously,
> ordinal costs would not make sense there).


FYI: routing protocols do have such node attribute. In ISIS we call=20
it "overload-bit" and allows an operator to define a router as a=20
non-transit device. Having the same capability in ALTO doesn't look=20
unreasonable to me.

I'd go a bit further and I believe having PID Capabilities TLV would
probably help.

s.


> Given this discussion and the confusion it caused, we'll have to go
> back and ensure this is explicitly stated in the draft.
>=20
> Thoughts?
> Rich
>=20
> On Mon, Jan 30, 2012 at 9:03 AM, Ben Niven-Jenkins
> <ben@niven-jenkins.co.uk> wrote:
>> Colleagues,
>>=20
>> In the current specification of ALTO, are costs always End to End?
>>=20
>> What I mean by that is when looking at ALTO cost maps is it possible =
to safely assume that of there is a cost between PIDX & PID Y and a cost =
between PIDY & PIDZ then the cost between PIDX & PIDZ can be calculated =
as cost(PIDX,PIDY)+cost(PIDY,PIDZ)? [If this assumption does hold, it is =
obviously not applicable to ordinal cost types].
>>=20
>> I suspect the answer is no, but I wanted to check what the definitive =
answer is.
>>=20
>>=20
>> For example if a cost map contains:
>>=20
>>   "map" : {
>>     "PID1": { "PID2": 1 },
>>     "PID2": { "PID1": 1, "PID3": 2 },
>>     "PID3": { "PID2": 2 }
>>=20
>> Can one assume that the cost between PID1 & PID3 is 3 (PID1->PID2 + =
PID2->PID3)?
>>=20
>> How about if the cost map contains:
>>=20
>>   "map" : {
>>     "PID1": { "PID2": 1, "PID3": 3 },
>>     "PID2": { "PID1": 1, "PID3": 2 },
>>     "PID3": { "PID2": 2, "PID1": 3 }
>>=20
>> Can one assume PID2 is on a path between PID1 & PID3?
>>=20
>> Thanks
>> Ben
>>=20
>> _______________________________________________
>> alto mailing list
>> alto@ietf.org
>> https://www.ietf.org/mailman/listinfo/alto
> _______________________________________________
> alto mailing list
> alto@ietf.org
> https://www.ietf.org/mailman/listinfo/alto
>=20


From richard.alimi@gmail.com  Wed Feb  1 02:17:18 2012
Return-Path: <richard.alimi@gmail.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 91AB521F8540 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 02:17:18 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.577
X-Spam-Level: 
X-Spam-Status: No, score=-2.577 tagged_above=-999 required=5 tests=[AWL=-0.200, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, J_CHICKENPOX_44=0.6, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xeIlT5IbfZEF for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 02:17:17 -0800 (PST)
Received: from mail-iy0-f172.google.com (mail-iy0-f172.google.com [209.85.210.172]) by ietfa.amsl.com (Postfix) with ESMTP id AA9DF21F84D9 for <alto@ietf.org>; Wed,  1 Feb 2012 02:17:17 -0800 (PST)
Received: by iagf6 with SMTP id f6so1614149iag.31 for <alto@ietf.org>; Wed, 01 Feb 2012 02:17:17 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:cc:content-type :content-transfer-encoding; bh=KyJ0KXdAkbjne31hwchkvRHi+sZoVnuniXN97E+OZFI=; b=wnjRQtuKWdVebOzfyTylyQ7yNV/T8G1Ooq4/vwqXm8hKtoEO/HZCb3W7txMNf2IiFM JdVbSSBnWnnkxIGTM4UCMb+nU9L7iXQj73Y532p0HkY3FQke8jOki/d++jS7BdhuFBrT Ia0pZZsWVqkXw1rmdV8mPcv61O3UG1nFYon/I=
Received: by 10.42.168.197 with SMTP id x5mr20898767icy.6.1328091437347; Wed, 01 Feb 2012 02:17:17 -0800 (PST)
MIME-Version: 1.0
Sender: richard.alimi@gmail.com
Received: by 10.231.69.7 with HTTP; Wed, 1 Feb 2012 02:16:57 -0800 (PST)
In-Reply-To: <1D8ECFB9-C14F-4661-9B68-F2DEE9FEA530@cisco.com>
References: <A35E0DC7-C133-480C-9967-EC62CAE9BCA0@niven-jenkins.co.uk> <CA+cvDaZGBS+WM9W=U_rf9qK6w+Ui=vrYTAp9UxKFb4JM+-=MTw@mail.gmail.com> <1D8ECFB9-C14F-4661-9B68-F2DEE9FEA530@cisco.com>
From: Richard Alimi <rich@velvetsea.net>
Date: Wed, 1 Feb 2012 02:16:57 -0800
X-Google-Sender-Auth: _KSLB7dKVf4P7LsbiNW3SypG-GU
Message-ID: <CA+cvDabQcOyw+w9SrRpBcSTS-8uX9HB+BDWL5Y5gFgSQNOrL+A@mail.gmail.com>
To: stefano previdi <sprevidi@cisco.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 10:17:18 -0000

On Wed, Feb 1, 2012 at 12:57 AM, stefano previdi <sprevidi@cisco.com> wrote=
:
> Hi Rich,
>
> On Feb 1, 2012, at 9:04 AM, Richard Alimi wrote:
>> I guess I'm a bit late to the party, so I can keep this simple and say
>> how the draft was currently intended:
>> (1) The choice of the term "endpoint" (those thingy's that are
>> contained in PIDs) was not an accident :)
>> (2) As such, each entry of the Cost Map is the *end-to-end* cost from
>> the source to the destination (I think this is what Ben meant by
>> "mesh"). =A0That is, the ALTO Server has already added up the costs on
>> the physical links along the path for you.
>
>
> do you intend that a cost map is supposed to be a full mesh of leaf
> nodes (as per analogy to SPF algorithm) ? IOW: you're not expected to
> transit across PIDs ?

Correct.  Transiting across PIDs would be an application-level concern
(e.g., constructing a P2P overlay) and outside of the scope of ALTO.
Since we're dealing with applications endpoints who are sending
resources back and forth (at least thats what ALTO was chartered for
:) ), the cost map indicates the cost of sending that resource from
one endpoint to another endpoint.

Now, in the CDN world, I agree those assumptions begin to change and
we start getting closer to an actual network topology (or aggregated
view of it)... hence this whole discussion.. :)

>
>> (3) If there is an entry missing from the Cost Map, it means the ALTO
>> Server doesn't know the cost from that source to destination (or isn't
>> willing to tell you). I guess you could generate a path, but the point
>> about whether a packet can actually traverse that path is a good one
>> -- ALTO doesn't make any claims on routing capabilities offered by
>> endpoints.
>>
>> Given that particular meaning, you could imagine doing overlay routing
>> on that (e.g., in a P2P swarm). In that sense, each endpoint is a node
>> on a graph, and you could imagine doing shortest path on that.
>> However, that's quite different than doing shortest path on the actual
>> network topology. =A0This was the reason for my comments a few times (I
>> guess not on the list, it was probably at the mic during a WG session)
>> that if we were to ever expose raw network topology via ALTO, then we
>> need to explicitly denote that it would be a different interpretation
>> of the Cost Map than we have today (an adjacency matrix seems like a
>> reasonable model here).
>
>
> I believe we have two orthogonal elements:
>
> 1. Information that can be leveraged by traditional SPF/Dijktra
> =A0 algorithm so to compute a topology using same algorithm/paradigms
> =A0 than the ones used in the network. This can be a cost map, I don't
> =A0 mind, as long as the semantic of the information is compatible with
> =A0 what we want to do with it. In this respect, Ben asked a valid
> =A0 question.

Yes - my only point is that if a cost map is to be interpreted this
way, it seems much cleaner/understandable to me if this mode is
explicitly indicated by the ALTO server.  More on this below...

> 2. The ability, for an ALTO server, to deliver a partial or aggregated
> =A0 or virtual topology that would NOT represent the reality of the
> =A0 network infrastructure but rather the topology that the application
> =A0 needs in order to optimize its business.

Right - I agree that is orthogonal.

>
> Implementors have worked quite a bit on the latter.
>
>
>> Along those lines, I'm a little bit hesitant of the proposal to add a
>> "transit" attribute to a PID, since that seems to conflate the two
>> possible interpretations of the cost map (end-to-end costs vs.
>> adjacency matrix). =A0If we wanted to convey the actual graph, how about
>> have an attribute for the entire cost map that indicates explicitly
>> that it is to be interpreted as an adjacency matrix (and obviously,
>> ordinal costs would not make sense there).
>
>
> FYI: routing protocols do have such node attribute. In ISIS we call
> it "overload-bit" and allows an operator to define a router as a
> non-transit device. Having the same capability in ALTO doesn't look
> unreasonable to me.
>
> I'd go a bit further and I believe having PID Capabilities TLV would
> probably help.

Okay, so let me try to put this all in perspective :) Imagine being an
application presented with a Cost Map from some ALTO Server.  Let's
assume for a minute that you don't have any pre-existing knowledge of
the ALTO Server and policies it uses to generate that information.  So
far on this thread we have discussed two possible ways to determine
the cost of sending some bytes from A -> B:
(1) Lookup (A, B) in the cost map [ this is what the protocol draft
currently prescribes ]
(2) Treat the cost map as an adjacency matrix, construct the
corresponding graph, compute the shortest path from A->B, and then
compute the cost based on that path (the particular mathematical
operation will depend on the cost type).

The application needs to know which algorithm to use.  What I'm
proposing is if an application is to interpret this as (2), then the
cost map returned by the ALTO Server be marked with an attribute
indicating that.

If when computing (2), certain PIDs were marked as non-transit, that
seems reasonable.  I would start to worry about scope-creep though.
There should be some dividing line to ensure we're not going to end up
building a generic data structure to carry what looks like a (perhaps
aggregated) link-state database.  In any case, thats probably a topic
for after (2) gets proposed/documented as an extension :)

The application of the non-transit PID attribute in (1) is less clear
to me, but I might imagine a few application-level heuristics that
could use that as input. For example, don't treat an endpoint in a
non-transit PID as a seed in a high-bandwidth peer in a P2P swarm.
However, it seems like those kinds of heuristics could equally be
gained by adjusting the costs in the Cost Map instead of introducing a
new attribute.

Thanks,
Rich

>
> s.
>
>
>> Given this discussion and the confusion it caused, we'll have to go
>> back and ensure this is explicitly stated in the draft.
>>
>> Thoughts?
>> Rich
>>
>> On Mon, Jan 30, 2012 at 9:03 AM, Ben Niven-Jenkins
>> <ben@niven-jenkins.co.uk> wrote:
>>> Colleagues,
>>>
>>> In the current specification of ALTO, are costs always End to End?
>>>
>>> What I mean by that is when looking at ALTO cost maps is it possible to=
 safely assume that of there is a cost between PIDX & PID Y and a cost betw=
een PIDY & PIDZ then the cost between PIDX & PIDZ can be calculated as cost=
(PIDX,PIDY)+cost(PIDY,PIDZ)? [If this assumption does hold, it is obviously=
 not applicable to ordinal cost types].
>>>
>>> I suspect the answer is no, but I wanted to check what the definitive a=
nswer is.
>>>
>>>
>>> For example if a cost map contains:
>>>
>>> =A0 "map" : {
>>> =A0 =A0 "PID1": { "PID2": 1 },
>>> =A0 =A0 "PID2": { "PID1": 1, "PID3": 2 },
>>> =A0 =A0 "PID3": { "PID2": 2 }
>>>
>>> Can one assume that the cost between PID1 & PID3 is 3 (PID1->PID2 + PID=
2->PID3)?
>>>
>>> How about if the cost map contains:
>>>
>>> =A0 "map" : {
>>> =A0 =A0 "PID1": { "PID2": 1, "PID3": 3 },
>>> =A0 =A0 "PID2": { "PID1": 1, "PID3": 2 },
>>> =A0 =A0 "PID3": { "PID2": 2, "PID1": 3 }
>>>
>>> Can one assume PID2 is on a path between PID1 & PID3?
>>>
>>> Thanks
>>> Ben
>>>
>>> _______________________________________________
>>> alto mailing list
>>> alto@ietf.org
>>> https://www.ietf.org/mailman/listinfo/alto
>> _______________________________________________
>> alto mailing list
>> alto@ietf.org
>> https://www.ietf.org/mailman/listinfo/alto
>>
>

From sprevidi@cisco.com  Wed Feb  1 02:56:18 2012
Return-Path: <sprevidi@cisco.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7A19021F8693 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 02:56:18 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.149
X-Spam-Level: 
X-Spam-Status: No, score=-102.149 tagged_above=-999 required=5 tests=[AWL=-0.150, BAYES_00=-2.599, J_CHICKENPOX_44=0.6, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9GOvKBEXglfm for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 02:56:17 -0800 (PST)
Received: from av-tac-bru.cisco.com (weird-brew.cisco.com [144.254.15.118]) by ietfa.amsl.com (Postfix) with ESMTP id CE33E21F8692 for <alto@ietf.org>; Wed,  1 Feb 2012 02:56:16 -0800 (PST)
X-TACSUNS: Virus Scanned
Received: from stew-brew.cisco.com (localhost.cisco.com [127.0.0.1]) by av-tac-bru.cisco.com (8.13.8+Sun/8.13.8) with ESMTP id q11AuDA9027011 for <alto@ietf.org>; Wed, 1 Feb 2012 11:56:13 +0100 (CET)
Received: from dhcp-10-55-92-12.cisco.com (dhcp-10-55-92-12.cisco.com [10.55.92.12]) by stew-brew.cisco.com (8.13.8+Sun/8.13.8) with ESMTP id q11AuBc7009415; Wed, 1 Feb 2012 11:56:12 +0100 (CET)
Mime-Version: 1.0 (Apple Message framework v1251.1)
Content-Type: text/plain; charset=iso-8859-1
From: stefano previdi <sprevidi@cisco.com>
In-Reply-To: <CA+cvDabQcOyw+w9SrRpBcSTS-8uX9HB+BDWL5Y5gFgSQNOrL+A@mail.gmail.com>
Date: Wed, 1 Feb 2012 11:56:03 +0100
Content-Transfer-Encoding: quoted-printable
Message-Id: <CF781218-9EA0-4786-AEAD-E1A4A485406D@cisco.com>
References: <A35E0DC7-C133-480C-9967-EC62CAE9BCA0@niven-jenkins.co.uk> <CA+cvDaZGBS+WM9W=U_rf9qK6w+Ui=vrYTAp9UxKFb4JM+-=MTw@mail.gmail.com> <1D8ECFB9-C14F-4661-9B68-F2DEE9FEA530@cisco.com> <CA+cvDabQcOyw+w9SrRpBcSTS-8uX9HB+BDWL5Y5gFgSQNOrL+A@mail.gmail.com>
To: Richard Alimi <rich@velvetsea.net>
X-Mailer: Apple Mail (2.1251.1)
Cc: alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 10:56:18 -0000

On Feb 1, 2012, at 11:16 AM, Richard Alimi wrote:

> On Wed, Feb 1, 2012 at 12:57 AM, stefano previdi <sprevidi@cisco.com> =
wrote:
>> Hi Rich,
>>=20
>> On Feb 1, 2012, at 9:04 AM, Richard Alimi wrote:
>>> I guess I'm a bit late to the party, so I can keep this simple and =
say
>>> how the draft was currently intended:
>>> (1) The choice of the term "endpoint" (those thingy's that are
>>> contained in PIDs) was not an accident :)
>>> (2) As such, each entry of the Cost Map is the *end-to-end* cost =
from
>>> the source to the destination (I think this is what Ben meant by
>>> "mesh").  That is, the ALTO Server has already added up the costs on
>>> the physical links along the path for you.
>>=20
>>=20
>> do you intend that a cost map is supposed to be a full mesh of leaf
>> nodes (as per analogy to SPF algorithm) ? IOW: you're not expected to
>> transit across PIDs ?
>=20
> Correct.  Transiting across PIDs would be an application-level concern
> (e.g., constructing a P2P overlay) and outside of the scope of ALTO.
> Since we're dealing with applications endpoints who are sending
> resources back and forth (at least thats what ALTO was chartered for
> :) ), the cost map indicates the cost of sending that resource from
> one endpoint to another endpoint.


yes, well, this was the original context and therefore your point is=20
right.


> Now, in the CDN world, I agree those assumptions begin to change and
> we start getting closer to an actual network topology (or aggregated
> view of it)... hence this whole discussion.. :)


exactly. And keep in mind that CDN is just another use case. Others=20
are being addressed by existing alto application as well (e.g.:=20
alto/topology guided DNS, Cloud Networking, ...).


>>> (3) If there is an entry missing from the Cost Map, it means the =
ALTO
>>> Server doesn't know the cost from that source to destination (or =
isn't
>>> willing to tell you). I guess you could generate a path, but the =
point
>>> about whether a packet can actually traverse that path is a good one
>>> -- ALTO doesn't make any claims on routing capabilities offered by
>>> endpoints.
>>>=20
>>> Given that particular meaning, you could imagine doing overlay =
routing
>>> on that (e.g., in a P2P swarm). In that sense, each endpoint is a =
node
>>> on a graph, and you could imagine doing shortest path on that.
>>> However, that's quite different than doing shortest path on the =
actual
>>> network topology.  This was the reason for my comments a few times =
(I
>>> guess not on the list, it was probably at the mic during a WG =
session)
>>> that if we were to ever expose raw network topology via ALTO, then =
we
>>> need to explicitly denote that it would be a different =
interpretation
>>> of the Cost Map than we have today (an adjacency matrix seems like a
>>> reasonable model here).
>>=20
>>=20
>> I believe we have two orthogonal elements:
>>=20
>> 1. Information that can be leveraged by traditional SPF/Dijktra
>>   algorithm so to compute a topology using same algorithm/paradigms
>>   than the ones used in the network. This can be a cost map, I don't
>>   mind, as long as the semantic of the information is compatible with
>>   what we want to do with it. In this respect, Ben asked a valid
>>   question.
>=20
> Yes - my only point is that if a cost map is to be interpreted this
> way, it seems much cleaner/understandable to me if this mode is
> explicitly indicated by the ALTO server.  More on this below...


Agreed, hence the capability/characteristics TLV.


>> 2. The ability, for an ALTO server, to deliver a partial or =
aggregated
>>   or virtual topology that would NOT represent the reality of the
>>   network infrastructure but rather the topology that the application
>>   needs in order to optimize its business.
>=20
> Right - I agree that is orthogonal.
>=20
>>=20
>> Implementors have worked quite a bit on the latter.
>>=20
>>=20
>>> Along those lines, I'm a little bit hesitant of the proposal to add =
a
>>> "transit" attribute to a PID, since that seems to conflate the two
>>> possible interpretations of the cost map (end-to-end costs vs.
>>> adjacency matrix).  If we wanted to convey the actual graph, how =
about
>>> have an attribute for the entire cost map that indicates explicitly
>>> that it is to be interpreted as an adjacency matrix (and obviously,
>>> ordinal costs would not make sense there).
>>=20
>>=20
>> FYI: routing protocols do have such node attribute. In ISIS we call
>> it "overload-bit" and allows an operator to define a router as a
>> non-transit device. Having the same capability in ALTO doesn't look
>> unreasonable to me.
>>=20
>> I'd go a bit further and I believe having PID Capabilities TLV would
>> probably help.
>=20
> Okay, so let me try to put this all in perspective :) Imagine being an
> application presented with a Cost Map from some ALTO Server.  Let's
> assume for a minute that you don't have any pre-existing knowledge of
> the ALTO Server and policies it uses to generate that information.


which is the case most of the time.


> So
> far on this thread we have discussed two possible ways to determine
> the cost of sending some bytes from A -> B:
> (1) Lookup (A, B) in the cost map [ this is what the protocol draft
> currently prescribes ]
> (2) Treat the cost map as an adjacency matrix, construct the
> corresponding graph, compute the shortest path from A->B, and then
> compute the cost based on that path (the particular mathematical
> operation will depend on the cost type).


just for completion, there's a 3rd way: ECS which allows an app not=20
to deal with mechanisms ALTO uses for building topology and doesn't
require the app to have any topology computational intelligence.


> The application needs to know which algorithm to use.  What I'm
> proposing is if an application is to interpret this as (2), then the
> cost map returned by the ALTO Server be marked with an attribute
> indicating that.


yes, I agree.=20


> If when computing (2), certain PIDs were marked as non-transit, that
> seems reasonable.  I would start to worry about scope-creep though.
> There should be some dividing line to ensure we're not going to end up
> building a generic data structure to carry what looks like a (perhaps
> aggregated) link-state database.  In any case, thats probably a topic
> for after (2) gets proposed/documented as an extension :)


so if I understand your point, we could have:
1. an attribute describing the whole map (graph vs. tree vs. mesh)
2. a pid attribute describing pid characteristics/capabilities

if so, then we are in agreement.


> The application of the non-transit PID attribute in (1) is less clear
> to me, but I might imagine a few application-level heuristics that
> could use that as input. For example, don't treat an endpoint in a
> non-transit PID as a seed in a high-bandwidth peer in a P2P swarm.


again, there are many other non-p2p apps out there.


> However, it seems like those kinds of heuristics could equally be
> gained by adjusting the costs in the Cost Map instead of introducing a
> new attribute.


You _MAY_ handle some cases with costs but it can become VERY complex=20
and also you MUST have a cost for each direction of the link/branch=20
connecting two PIDs.

As an implementor, I'd discourage that way...


s.


>=20
> Thanks,
> Rich
>=20
>>=20
>> s.
>>=20
>>=20
>>> Given this discussion and the confusion it caused, we'll have to go
>>> back and ensure this is explicitly stated in the draft.
>>>=20
>>> Thoughts?
>>> Rich
>>>=20
>>> On Mon, Jan 30, 2012 at 9:03 AM, Ben Niven-Jenkins
>>> <ben@niven-jenkins.co.uk> wrote:
>>>> Colleagues,
>>>>=20
>>>> In the current specification of ALTO, are costs always End to End?
>>>>=20
>>>> What I mean by that is when looking at ALTO cost maps is it =
possible to safely assume that of there is a cost between PIDX & PID Y =
and a cost between PIDY & PIDZ then the cost between PIDX & PIDZ can be =
calculated as cost(PIDX,PIDY)+cost(PIDY,PIDZ)? [If this assumption does =
hold, it is obviously not applicable to ordinal cost types].
>>>>=20
>>>> I suspect the answer is no, but I wanted to check what the =
definitive answer is.
>>>>=20
>>>>=20
>>>> For example if a cost map contains:
>>>>=20
>>>>   "map" : {
>>>>     "PID1": { "PID2": 1 },
>>>>     "PID2": { "PID1": 1, "PID3": 2 },
>>>>     "PID3": { "PID2": 2 }
>>>>=20
>>>> Can one assume that the cost between PID1 & PID3 is 3 (PID1->PID2 + =
PID2->PID3)?
>>>>=20
>>>> How about if the cost map contains:
>>>>=20
>>>>   "map" : {
>>>>     "PID1": { "PID2": 1, "PID3": 3 },
>>>>     "PID2": { "PID1": 1, "PID3": 2 },
>>>>     "PID3": { "PID2": 2, "PID1": 3 }
>>>>=20
>>>> Can one assume PID2 is on a path between PID1 & PID3?
>>>>=20
>>>> Thanks
>>>> Ben
>>>>=20
>>>> _______________________________________________
>>>> alto mailing list
>>>> alto@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/alto
>>> _______________________________________________
>>> alto mailing list
>>> alto@ietf.org
>>> https://www.ietf.org/mailman/listinfo/alto
>>>=20
>>=20
>=20


From w.roome@alcatel-lucent.com  Wed Feb  1 07:08:15 2012
Return-Path: <w.roome@alcatel-lucent.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E7A6411E80D1 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 07:08:14 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -9.999
X-Spam-Level: 
X-Spam-Status: No, score=-9.999 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, J_CHICKENPOX_44=0.6, RCVD_IN_DNSWL_HI=-8]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id oBPtOtBi0aeJ for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 07:08:14 -0800 (PST)
Received: from ihemail1.lucent.com (ihemail1.lucent.com [135.245.0.33]) by ietfa.amsl.com (Postfix) with ESMTP id B478511E8073 for <alto@ietf.org>; Wed,  1 Feb 2012 07:08:07 -0800 (PST)
Received: from usnavsmail2.ndc.alcatel-lucent.com (usnavsmail2.ndc.alcatel-lucent.com [135.3.39.10]) by ihemail1.lucent.com (8.13.8/IER-o) with ESMTP id q11F87nG006020 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for <alto@ietf.org>; Wed, 1 Feb 2012 09:08:07 -0600 (CST)
Received: from umail.lucent.com (umail-ce2.ndc.lucent.com [135.3.40.63]) by usnavsmail2.ndc.alcatel-lucent.com (8.14.3/8.14.3/GMO) with ESMTP id q11F86LO022722 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for <alto@ietf.org>; Wed, 1 Feb 2012 09:08:07 -0600
Received: from [135.222.104.84] (wdr-i7mbp.mh.lucent.com [135.222.104.84]) by umail.lucent.com (8.13.8/TPES) with ESMTP id q11F7wj6026656 for <alto@ietf.org>; Wed, 1 Feb 2012 09:08:06 -0600 (CST)
User-Agent: Microsoft-MacOutlook/14.14.0.111121
Date: Wed, 01 Feb 2012 10:07:58 -0500
From: Bill Roome <w.roome@alcatel-lucent.com>
To: <alto@ietf.org>
Message-ID: <CB4EBC77.A75E%w.roome@alcatel-lucent.com>
Thread-Topic: Are ALTO costs always End to End?
In-Reply-To: <mailman.4534.1328093779.3200.alto@ietf.org>
Mime-version: 1.0
Content-type: text/plain; charset="US-ASCII"
Content-transfer-encoding: 7bit
X-Scanned-By: MIMEDefang 2.57 on 135.245.2.33
X-Scanned-By: MIMEDefang 2.64 on 135.3.39.10
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 15:08:15 -0000

I guess I'm also late to this discussion. But I always assumed:

1. The ALTO costs are end-to-end. Always.

2. An ALTO server is expected to give the costs between *every* pair of
PIDs. It can omit a cost only if it cannot determine that cost at all.

3. You cannot infer a topology from the 'holes' in the cost map.

4. If it were possible to calculate the missing costs by using SPF (or
whatever) on the other costs, then the ALTO server should do that! The
ALTO server shouldn't push that task onto the clients.

	- Bill Roome


>On Mon, Jan 30, 2012 at 9:03 AM, Ben Niven-Jenkins
><ben@niven-jenkins.co.uk> wrote:
>>Colleagues,
>>
>>In the current specification of ALTO, are costs always End to End?
>>
>>What I mean by that is when looking at ALTO cost maps is it possible to
>>safely assume that of there is a cost between PIDX & PID Y and a cost
>>between PIDY & PIDZ then the cost between PIDX & PIDZ can be calculated
>>as cost(PIDX,PIDY)+cost(PIDY,PIDZ)? [If this assumption does hold, it is
>>obviously not applicable to ordinal cost types].
>>
>>I suspect the answer is no, but I wanted to check what the definitive
>>answer is.
>>
>>
>>For example if a cost map contains:
>>
>>? "map" : {
>>? ? "PID1": { "PID2": 1 },
>>? ? "PID2": { "PID1": 1, "PID3": 2 },
>>? ? "PID3": { "PID2": 2 }
>>
>>Can one assume that the cost between PID1 & PID3 is 3 (PID1->PID2 +
>>PID2->PID3)?
>>
>>How about if the cost map contains:
>>
>>? "map" : {
>>? ? "PID1": { "PID2": 1, "PID3": 3 },
>>? ? "PID2": { "PID1": 1, "PID3": 2 },
>>? ? "PID3": { "PID2": 2, "PID1": 3 }
>>
>>Can one assume PID2 is on a path between PID1 & PID3?
>>
>>Thanks
>>Ben
>>
>>_______________________________________________
>>alto mailing list
>>alto@ietf.org
>>https://www.ietf.org/mailman/listinfo/alto
>



From sebi@gw01.ehlo.wurstkaes.de  Wed Feb  1 08:50:47 2012
Return-Path: <sebi@gw01.ehlo.wurstkaes.de>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 72D1C11E8074 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 08:50:47 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.638
X-Spam-Level: 
X-Spam-Status: No, score=-1.638 tagged_above=-999 required=5 tests=[AWL=0.612,  BAYES_00=-2.599, HELO_EQ_DE=0.35]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RWU5yGvIKE06 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 08:50:47 -0800 (PST)
Received: from gw01.ehlo.wurstkaes.de (gw01.ehlo.wurstkaes.de [188.246.4.151]) by ietfa.amsl.com (Postfix) with ESMTP id C5FB511E8071 for <alto@ietf.org>; Wed,  1 Feb 2012 08:50:46 -0800 (PST)
Received: from sebi by gw01.ehlo.wurstkaes.de with local (Exim 4.72) (envelope-from <sebi@gw01.ehlo.wurstkaes.de>) id 1RsdOD-0005hX-2I; Wed, 01 Feb 2012 17:50:33 +0100
Date: Wed, 1 Feb 2012 17:50:33 +0100
From: Sebastian Kiesel <ietf-alto@skiesel.de>
To: Bill Roome <w.roome@alcatel-lucent.com>
Message-ID: <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de>
References: <mailman.4534.1328093779.3200.alto@ietf.org> <CB4EBC77.A75E%w.roome@alcatel-lucent.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CB4EBC77.A75E%w.roome@alcatel-lucent.com>
Accept-Languages: en, de
Organization: my personal mail account
User-Agent: Mutt/1.5.20 (2009-06-14)
Cc: alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 16:50:47 -0000

Bill, all,

On Wed, Feb 01, 2012 at 10:07:58AM -0500, Bill Roome wrote:
> I guess I'm also late to this discussion. But I always assumed:
> 
> 1. The ALTO costs are end-to-end. Always.

ack.

> 2. An ALTO server is expected to give the costs between *every* pair of
> PIDs. It can omit a cost only if it cannot determine that cost at all.

My assumption was different: I'd say that the protocol specification
must allow the ALTO server to return a full NxN matrix indicating all
end-to-end costs between all N PIDs (maybe even different costs in the
two directions between two given PIDs).  

But I also assume that in a real deployment we will rather have a very
sparse matrix, with the default value "I don't know / I am not allowed
to tell". I could even imagine that my ISP would operate an ALTO server,
which gives my hosts only an 1xN vector, which tells me how "expensive"
it is for me to reach all possible destinations in the Internet.

> 3. You cannot infer a topology from the 'holes' in the cost map.

ack.

> 4. If it were possible to calculate the missing costs by using SPF (or
> whatever) on the other costs, then the ALTO server should do that! The
> ALTO server shouldn't push that task onto the clients.

ack.


Btw: which application needs to know the topology? If I remember
correctly once upon a time our primary goal was assisting peer selection
(i.e., I know that I can get the desired resource form IP address A, B,
or C, so please ALTO service tell me which address I should give a
connect()-try first). To that end, the 1xN vector would be enough.

  -- Sebastian

From vkg@bell-labs.com  Wed Feb  1 08:55:59 2012
Return-Path: <vkg@bell-labs.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7470611E80BA for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 08:55:59 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -106.599
X-Spam-Level: 
X-Spam-Status: No, score=-106.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1fsgRl8VyxPa for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 08:55:58 -0800 (PST)
Received: from ihemail4.lucent.com (ihemail4.lucent.com [135.245.0.39]) by ietfa.amsl.com (Postfix) with ESMTP id B11FF11E80B2 for <alto@ietf.org>; Wed,  1 Feb 2012 08:55:58 -0800 (PST)
Received: from usnavsmail1.ndc.alcatel-lucent.com (usnavsmail1.ndc.alcatel-lucent.com [135.3.39.9]) by ihemail4.lucent.com (8.13.8/IER-o) with ESMTP id q11GtvJM010347 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for <alto@ietf.org>; Wed, 1 Feb 2012 10:55:57 -0600 (CST)
Received: from umail.lucent.com (umail-ce2.ndc.lucent.com [135.3.40.63]) by usnavsmail1.ndc.alcatel-lucent.com (8.14.3/8.14.3/GMO) with ESMTP id q11GtuTe030413 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for <alto@ietf.org>; Wed, 1 Feb 2012 10:55:57 -0600
Received: from shoonya.ih.lucent.com (vkg.lra.lucent.com [135.244.42.252]) by umail.lucent.com (8.13.8/TPES) with ESMTP id q11GtuIO021364 for <alto@ietf.org>; Wed, 1 Feb 2012 10:55:56 -0600 (CST)
Message-ID: <4F296F97.8000002@bell-labs.com>
Date: Wed, 01 Feb 2012 11:00:07 -0600
From: "Vijay K. Gurbani" <vkg@bell-labs.com>
Organization: Bell Laboratories, Alcatel-Lucent
User-Agent: Mozilla/5.0 (X11; Linux i686; rv:9.0) Gecko/20111222 Thunderbird/9.0
MIME-Version: 1.0
To: alto@ietf.org
References: <mailman.4534.1328093779.3200.alto@ietf.org> <CB4EBC77.A75E%w.roome@alcatel-lucent.com> <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de>
In-Reply-To: <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-Scanned-By: MIMEDefang 2.57 on 135.245.2.39
X-Scanned-By: MIMEDefang 2.64 on 135.3.39.9
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 16:55:59 -0000

On 02/01/2012 10:50 AM, Sebastian Kiesel wrote:
> Btw: which application needs to know the topology? If I remember
> correctly once upon a time our primary goal was assisting peer
> selection (i.e., I know that I can get the desired resource form IP
> address A, B, or C, so please ALTO service tell me which address I
> should give a connect()-try first). To that end, the 1xN vector would
> be enough.

Sebastian: There are other emerging uses of ALTO being discussed.  CDNs
and data centers are the two widely cited ones.  See [1] for more on
this.

[1] http://tools.ietf.org/html/draft-marocco-alto-next-00

Thanks,

- vijay
-- 
Vijay K. Gurbani, Bell Laboratories, Alcatel-Lucent
1960 Lucent Lane, Rm. 9C-533, Naperville, Illinois 60566 (USA)
Email: vkg@{bell-labs.com,acm.org} / vijay.gurbani@alcatel-lucent.com
Web:   http://ect.bell-labs.com/who/vkg/

From sprevidi@cisco.com  Wed Feb  1 09:09:42 2012
Return-Path: <sprevidi@cisco.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CF42411E8120 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 09:09:42 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.419
X-Spam-Level: 
X-Spam-Status: No, score=-102.419 tagged_above=-999 required=5 tests=[AWL=0.180, BAYES_00=-2.599, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 62L9pJvzqxPL for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 09:09:41 -0800 (PST)
Received: from av-tac-bru.cisco.com (weird-brew.cisco.com [144.254.15.118]) by ietfa.amsl.com (Postfix) with ESMTP id 6ACEA11E8123 for <alto@ietf.org>; Wed,  1 Feb 2012 09:09:41 -0800 (PST)
X-TACSUNS: Virus Scanned
Received: from stew-brew.cisco.com (localhost.cisco.com [127.0.0.1]) by av-tac-bru.cisco.com (8.13.8+Sun/8.13.8) with ESMTP id q11H4OJB011728 for <alto@ietf.org>; Wed, 1 Feb 2012 18:04:24 +0100 (CET)
Received: from dhcp-10-55-82-59.cisco.com (dhcp-10-55-82-59.cisco.com [10.55.82.59]) by stew-brew.cisco.com (8.13.8+Sun/8.13.8) with ESMTP id q11H4NbZ004558; Wed, 1 Feb 2012 18:04:23 +0100 (CET)
Mime-Version: 1.0 (Apple Message framework v1251.1)
Content-Type: text/plain; charset=us-ascii
From: stefano previdi <sprevidi@cisco.com>
In-Reply-To: <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de>
Date: Wed, 1 Feb 2012 18:04:23 +0100
Content-Transfer-Encoding: 7bit
Message-Id: <73FCC4FE-5CEE-43A9-B8F1-8BD45D5DFFDB@cisco.com>
References: <mailman.4534.1328093779.3200.alto@ietf.org> <CB4EBC77.A75E%w.roome@alcatel-lucent.com> <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de>
To: Sebastian Kiesel <ietf-alto@skiesel.de>
X-Mailer: Apple Mail (2.1251.1)
Cc: Bill Roome <w.roome@alcatel-lucent.com>, alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 17:09:43 -0000

On Feb 1, 2012, at 5:50 PM, Sebastian Kiesel wrote:

> Bill, all,
> 
> On Wed, Feb 01, 2012 at 10:07:58AM -0500, Bill Roome wrote:
>> I guess I'm also late to this discussion. But I always assumed:
>> 
>> 1. The ALTO costs are end-to-end. Always.
> 
> ack.
> 
>> 2. An ALTO server is expected to give the costs between *every* pair of
>> PIDs. It can omit a cost only if it cannot determine that cost at all.
> 
> My assumption was different:


the differences in assumptions tend to demonstrate we would need 
to be more precise on the semantic.


> I'd say that the protocol specification
> must allow the ALTO server to return a full NxN matrix indicating all
> end-to-end costs between all N PIDs (maybe even different costs in the
> two directions between two given PIDs).  
> 
> But I also assume that in a real deployment we will rather have a very
> sparse matrix, with the default value "I don't know / I am not allowed
> to tell". I could even imagine that my ISP would operate an ALTO server,
> which gives my hosts only an 1xN vector, which tells me how "expensive"
> it is for me to reach all possible destinations in the Internet.
> 
>> 3. You cannot infer a topology from the 'holes' in the cost map.
> 
> ack.
> 
>> 4. If it were possible to calculate the missing costs by using SPF (or
>> whatever) on the other costs, then the ALTO server should do that! The
>> ALTO server shouldn't push that task onto the clients.
> 
> ack.
> 
> 
> Btw: which application needs to know the topology?


E.g.: a CDN may want to operate autonomously with maps rather than 
using ECS. Not that I'm advocating this but it all depends on where 
do you want to place the complexity of topology hint: inside ALTO 
server (ECS like of approach) or inside the application (Maps that 
allows you to compute paths/trees).


> If I remember
> correctly once upon a time our primary goal was assisting peer selection
> (i.e., I know that I can get the desired resource form IP address A, B,
> or C, so please ALTO service tell me which address I should give a
> connect()-try first). To that end, the 1xN vector would be enough.


The CDN is not substantially different in its logic.

s.


> 
>  -- Sebastian
> _______________________________________________
> alto mailing list
> alto@ietf.org
> https://www.ietf.org/mailman/listinfo/alto
> 


From richard.alimi@gmail.com  Wed Feb  1 09:26:33 2012
Return-Path: <richard.alimi@gmail.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C71E921F87F5 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 09:26:33 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.527
X-Spam-Level: 
X-Spam-Status: No, score=-2.527 tagged_above=-999 required=5 tests=[AWL=-0.150, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, J_CHICKENPOX_44=0.6, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 34ZihPM0wJn0 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 09:26:33 -0800 (PST)
Received: from mail-gy0-f172.google.com (mail-gy0-f172.google.com [209.85.160.172]) by ietfa.amsl.com (Postfix) with ESMTP id C53AE21F87F4 for <alto@ietf.org>; Wed,  1 Feb 2012 09:26:32 -0800 (PST)
Received: by ghbg16 with SMTP id g16so803967ghb.31 for <alto@ietf.org>; Wed, 01 Feb 2012 09:26:32 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:cc:content-type :content-transfer-encoding; bh=IJHMR6kCH0KWDf6MnBTLbwMP7rLqRHjbcXAz9rD3cwg=; b=tUXxKRr+EuxHNi2hejxC/RuW8RkyBE4vn1TneLRuNFTPezBgxskm8BNIyatjwqchr5 a4QMEffOYBo4vMDe2cAr0dIwmyoZRIdxacJviKGBIY2UR9IXy0rbLMDOiULrxGlTPgpq Nj1btY8cXVBC+bq1Pth6DwXtlNVQpe/iPKGWE=
Received: by 10.50.189.134 with SMTP id gi6mr8231930igc.18.1328117192170; Wed, 01 Feb 2012 09:26:32 -0800 (PST)
MIME-Version: 1.0
Sender: richard.alimi@gmail.com
Received: by 10.231.69.7 with HTTP; Wed, 1 Feb 2012 09:26:12 -0800 (PST)
In-Reply-To: <CF781218-9EA0-4786-AEAD-E1A4A485406D@cisco.com>
References: <A35E0DC7-C133-480C-9967-EC62CAE9BCA0@niven-jenkins.co.uk> <CA+cvDaZGBS+WM9W=U_rf9qK6w+Ui=vrYTAp9UxKFb4JM+-=MTw@mail.gmail.com> <1D8ECFB9-C14F-4661-9B68-F2DEE9FEA530@cisco.com> <CA+cvDabQcOyw+w9SrRpBcSTS-8uX9HB+BDWL5Y5gFgSQNOrL+A@mail.gmail.com> <CF781218-9EA0-4786-AEAD-E1A4A485406D@cisco.com>
From: Richard Alimi <rich@velvetsea.net>
Date: Wed, 1 Feb 2012 09:26:12 -0800
X-Google-Sender-Auth: RsSIxN36luTq_YjDltv7uyinO68
Message-ID: <CA+cvDabBQmy6UKVKLT+ibpUPqiKW36dOTtfwQ5DETef1BXQ7DQ@mail.gmail.com>
To: stefano previdi <sprevidi@cisco.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 17:26:33 -0000

On Wed, Feb 1, 2012 at 2:56 AM, stefano previdi <sprevidi@cisco.com> wrote:
>
> On Feb 1, 2012, at 11:16 AM, Richard Alimi wrote:
>
>> On Wed, Feb 1, 2012 at 12:57 AM, stefano previdi <sprevidi@cisco.com> wr=
ote:
>>> Hi Rich,
>>>
>>> On Feb 1, 2012, at 9:04 AM, Richard Alimi wrote:
>>>> I guess I'm a bit late to the party, so I can keep this simple and say
>>>> how the draft was currently intended:
>>>> (1) The choice of the term "endpoint" (those thingy's that are
>>>> contained in PIDs) was not an accident :)
>>>> (2) As such, each entry of the Cost Map is the *end-to-end* cost from
>>>> the source to the destination (I think this is what Ben meant by
>>>> "mesh"). =A0That is, the ALTO Server has already added up the costs on
>>>> the physical links along the path for you.
>>>
>>>
>>> do you intend that a cost map is supposed to be a full mesh of leaf
>>> nodes (as per analogy to SPF algorithm) ? IOW: you're not expected to
>>> transit across PIDs ?
>>
>> Correct. =A0Transiting across PIDs would be an application-level concern
>> (e.g., constructing a P2P overlay) and outside of the scope of ALTO.
>> Since we're dealing with applications endpoints who are sending
>> resources back and forth (at least thats what ALTO was chartered for
>> :) ), the cost map indicates the cost of sending that resource from
>> one endpoint to another endpoint.
>
>
> yes, well, this was the original context and therefore your point is
> right.
>
>
>> Now, in the CDN world, I agree those assumptions begin to change and
>> we start getting closer to an actual network topology (or aggregated
>> view of it)... hence this whole discussion.. :)
>
>
> exactly. And keep in mind that CDN is just another use case. Others
> are being addressed by existing alto application as well (e.g.:
> alto/topology guided DNS, Cloud Networking, ...).
>
>
>>>> (3) If there is an entry missing from the Cost Map, it means the ALTO
>>>> Server doesn't know the cost from that source to destination (or isn't
>>>> willing to tell you). I guess you could generate a path, but the point
>>>> about whether a packet can actually traverse that path is a good one
>>>> -- ALTO doesn't make any claims on routing capabilities offered by
>>>> endpoints.
>>>>
>>>> Given that particular meaning, you could imagine doing overlay routing
>>>> on that (e.g., in a P2P swarm). In that sense, each endpoint is a node
>>>> on a graph, and you could imagine doing shortest path on that.
>>>> However, that's quite different than doing shortest path on the actual
>>>> network topology. =A0This was the reason for my comments a few times (=
I
>>>> guess not on the list, it was probably at the mic during a WG session)
>>>> that if we were to ever expose raw network topology via ALTO, then we
>>>> need to explicitly denote that it would be a different interpretation
>>>> of the Cost Map than we have today (an adjacency matrix seems like a
>>>> reasonable model here).
>>>
>>>
>>> I believe we have two orthogonal elements:
>>>
>>> 1. Information that can be leveraged by traditional SPF/Dijktra
>>> =A0 algorithm so to compute a topology using same algorithm/paradigms
>>> =A0 than the ones used in the network. This can be a cost map, I don't
>>> =A0 mind, as long as the semantic of the information is compatible with
>>> =A0 what we want to do with it. In this respect, Ben asked a valid
>>> =A0 question.
>>
>> Yes - my only point is that if a cost map is to be interpreted this
>> way, it seems much cleaner/understandable to me if this mode is
>> explicitly indicated by the ALTO server. =A0More on this below...
>
>
> Agreed, hence the capability/characteristics TLV.
>
>
>>> 2. The ability, for an ALTO server, to deliver a partial or aggregated
>>> =A0 or virtual topology that would NOT represent the reality of the
>>> =A0 network infrastructure but rather the topology that the application
>>> =A0 needs in order to optimize its business.
>>
>> Right - I agree that is orthogonal.
>>
>>>
>>> Implementors have worked quite a bit on the latter.
>>>
>>>
>>>> Along those lines, I'm a little bit hesitant of the proposal to add a
>>>> "transit" attribute to a PID, since that seems to conflate the two
>>>> possible interpretations of the cost map (end-to-end costs vs.
>>>> adjacency matrix). =A0If we wanted to convey the actual graph, how abo=
ut
>>>> have an attribute for the entire cost map that indicates explicitly
>>>> that it is to be interpreted as an adjacency matrix (and obviously,
>>>> ordinal costs would not make sense there).
>>>
>>>
>>> FYI: routing protocols do have such node attribute. In ISIS we call
>>> it "overload-bit" and allows an operator to define a router as a
>>> non-transit device. Having the same capability in ALTO doesn't look
>>> unreasonable to me.
>>>
>>> I'd go a bit further and I believe having PID Capabilities TLV would
>>> probably help.
>>
>> Okay, so let me try to put this all in perspective :) Imagine being an
>> application presented with a Cost Map from some ALTO Server. =A0Let's
>> assume for a minute that you don't have any pre-existing knowledge of
>> the ALTO Server and policies it uses to generate that information.
>
>
> which is the case most of the time.
>
>
>> So
>> far on this thread we have discussed two possible ways to determine
>> the cost of sending some bytes from A -> B:
>> (1) Lookup (A, B) in the cost map [ this is what the protocol draft
>> currently prescribes ]
>> (2) Treat the cost map as an adjacency matrix, construct the
>> corresponding graph, compute the shortest path from A->B, and then
>> compute the cost based on that path (the particular mathematical
>> operation will depend on the cost type).
>
>
> just for completion, there's a 3rd way: ECS which allows an app not
> to deal with mechanisms ALTO uses for building topology and doesn't
> require the app to have any topology computational intelligence.
>
>
>> The application needs to know which algorithm to use. =A0What I'm
>> proposing is if an application is to interpret this as (2), then the
>> cost map returned by the ALTO Server be marked with an attribute
>> indicating that.
>
>
> yes, I agree.
>
>
>> If when computing (2), certain PIDs were marked as non-transit, that
>> seems reasonable. =A0I would start to worry about scope-creep though.
>> There should be some dividing line to ensure we're not going to end up
>> building a generic data structure to carry what looks like a (perhaps
>> aggregated) link-state database. =A0In any case, thats probably a topic
>> for after (2) gets proposed/documented as an extension :)
>
>
> so if I understand your point, we could have:
> 1. an attribute describing the whole map (graph vs. tree vs. mesh)
> 2. a pid attribute describing pid characteristics/capabilities
>
> if so, then we are in agreement.

Yes.  I believe we are in agreement.

>
>
>> The application of the non-transit PID attribute in (1) is less clear
>> to me, but I might imagine a few application-level heuristics that
>> could use that as input. For example, don't treat an endpoint in a
>> non-transit PID as a seed in a high-bandwidth peer in a P2P swarm.
>
>
> again, there are many other non-p2p apps out there.
>
>
>> However, it seems like those kinds of heuristics could equally be
>> gained by adjusting the costs in the Cost Map instead of introducing a
>> new attribute.
>
>
> You _MAY_ handle some cases with costs but it can become VERY complex
> and also you MUST have a cost for each direction of the link/branch
> connecting two PIDs.
>
> As an implementor, I'd discourage that way...

That's true.

Rich

>
>
> s.
>
>
>>
>> Thanks,
>> Rich
>>
>>>
>>> s.
>>>
>>>
>>>> Given this discussion and the confusion it caused, we'll have to go
>>>> back and ensure this is explicitly stated in the draft.
>>>>
>>>> Thoughts?
>>>> Rich
>>>>
>>>> On Mon, Jan 30, 2012 at 9:03 AM, Ben Niven-Jenkins
>>>> <ben@niven-jenkins.co.uk> wrote:
>>>>> Colleagues,
>>>>>
>>>>> In the current specification of ALTO, are costs always End to End?
>>>>>
>>>>> What I mean by that is when looking at ALTO cost maps is it possible =
to safely assume that of there is a cost between PIDX & PID Y and a cost be=
tween PIDY & PIDZ then the cost between PIDX & PIDZ can be calculated as co=
st(PIDX,PIDY)+cost(PIDY,PIDZ)? [If this assumption does hold, it is obvious=
ly not applicable to ordinal cost types].
>>>>>
>>>>> I suspect the answer is no, but I wanted to check what the definitive=
 answer is.
>>>>>
>>>>>
>>>>> For example if a cost map contains:
>>>>>
>>>>> =A0 "map" : {
>>>>> =A0 =A0 "PID1": { "PID2": 1 },
>>>>> =A0 =A0 "PID2": { "PID1": 1, "PID3": 2 },
>>>>> =A0 =A0 "PID3": { "PID2": 2 }
>>>>>
>>>>> Can one assume that the cost between PID1 & PID3 is 3 (PID1->PID2 + P=
ID2->PID3)?
>>>>>
>>>>> How about if the cost map contains:
>>>>>
>>>>> =A0 "map" : {
>>>>> =A0 =A0 "PID1": { "PID2": 1, "PID3": 3 },
>>>>> =A0 =A0 "PID2": { "PID1": 1, "PID3": 2 },
>>>>> =A0 =A0 "PID3": { "PID2": 2, "PID1": 3 }
>>>>>
>>>>> Can one assume PID2 is on a path between PID1 & PID3?
>>>>>
>>>>> Thanks
>>>>> Ben
>>>>>
>>>>> _______________________________________________
>>>>> alto mailing list
>>>>> alto@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/alto
>>>> _______________________________________________
>>>> alto mailing list
>>>> alto@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/alto
>>>>
>>>
>>
>

From richard.alimi@gmail.com  Wed Feb  1 09:33:24 2012
Return-Path: <richard.alimi@gmail.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 28F1611E8144 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 09:33:24 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.797
X-Spam-Level: 
X-Spam-Status: No, score=-2.797 tagged_above=-999 required=5 tests=[AWL=0.180,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NRXpIwLkTMJv for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 09:33:23 -0800 (PST)
Received: from mail-iy0-f172.google.com (mail-iy0-f172.google.com [209.85.210.172]) by ietfa.amsl.com (Postfix) with ESMTP id 72B2611E811D for <alto@ietf.org>; Wed,  1 Feb 2012 09:33:23 -0800 (PST)
Received: by iagf6 with SMTP id f6so2217753iag.31 for <alto@ietf.org>; Wed, 01 Feb 2012 09:33:23 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:cc:content-type :content-transfer-encoding; bh=CXOCgrI8syhx/EjQLxdyigtzvkoY317DgaXKuuihPo8=; b=JYw9yuwsFlBrJLlpSfkO8M6RCmtQTPI9renVeT/aD5xlB1g5S8x2vENezc6IJi9y1p hPZfsWrT4ZNfB6YxRAvechJY9gK4FZc93tniecK8PhsIJPGVTz32UHD8YzZdYhb5ms3x 1EERLz9COSobzsR5BwXl0VHdxElJVJ0sVR2SU=
Received: by 10.42.153.201 with SMTP id n9mr6411858icw.14.1328117603154; Wed, 01 Feb 2012 09:33:23 -0800 (PST)
MIME-Version: 1.0
Sender: richard.alimi@gmail.com
Received: by 10.231.69.7 with HTTP; Wed, 1 Feb 2012 09:33:03 -0800 (PST)
In-Reply-To: <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de>
References: <mailman.4534.1328093779.3200.alto@ietf.org> <CB4EBC77.A75E%w.roome@alcatel-lucent.com> <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de>
From: Richard Alimi <rich@velvetsea.net>
Date: Wed, 1 Feb 2012 09:33:03 -0800
X-Google-Sender-Auth: KKcrfCJHlvMgx7WsD1Lq4i6ufwY
Message-ID: <CA+cvDaY2XAbQ8FbZuitwdoxQ57_QLxM2GyihVi5FerRdNG-sdQ@mail.gmail.com>
To: Sebastian Kiesel <ietf-alto@skiesel.de>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: Bill Roome <w.roome@alcatel-lucent.com>, alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 17:33:24 -0000

On Wed, Feb 1, 2012 at 8:50 AM, Sebastian Kiesel <ietf-alto@skiesel.de> wro=
te:
> Bill, all,
>
> On Wed, Feb 01, 2012 at 10:07:58AM -0500, Bill Roome wrote:
>> I guess I'm also late to this discussion. But I always assumed:
>>
>> 1. The ALTO costs are end-to-end. Always.
>
> ack.
>
>> 2. An ALTO server is expected to give the costs between *every* pair of
>> PIDs. It can omit a cost only if it cannot determine that cost at all.
>
> My assumption was different: I'd say that the protocol specification
> must allow the ALTO server to return a full NxN matrix indicating all
> end-to-end costs between all N PIDs (maybe even different costs in the
> two directions between two given PIDs).
>
> But I also assume that in a real deployment we will rather have a very
> sparse matrix, with the default value "I don't know / I am not allowed
> to tell". I could even imagine that my ISP would operate an ALTO server,
> which gives my hosts only an 1xN vector, which tells me how "expensive"
> it is for me to reach all possible destinations in the Internet.

The current interpretation is: The ALTO protocol makes no statement on
what entries in the cost map must or must not be filled in.  It is
assumed that if an entry is not there, the ALTO server did not wish to
make a statement on the cost (either because it didn't know, or
because it didn't want to tell you).  IIRC, Vijay proposed an
extension that would provide some more detail there, which was to
indicate how certain the ALTO Server was in its answer.

Certainly let us know if you think the protocol draft needs to make
this clearer.

>
>> 3. You cannot infer a topology from the 'holes' in the cost map.
>
> ack.
>
>> 4. If it were possible to calculate the missing costs by using SPF (or
>> whatever) on the other costs, then the ALTO server should do that! The
>> ALTO server shouldn't push that task onto the clients.
>
> ack.
>
>
> Btw: which application needs to know the topology? If I remember
> correctly once upon a time our primary goal was assisting peer selection
> (i.e., I know that I can get the desired resource form IP address A, B,
> or C, so please ALTO service tell me which address I should give a
> connect()-try first). To that end, the 1xN vector would be enough.

In the CDN world (and other non-P2P cases) in some deployements there
may be additional trust between the ALTO Server and ALTO Client.  That
opens up the possibility of exposing more fine-grained information,
including something resembling the topology of the network.  Yes, it
is straying from the original context of P2P - but the charter now is
open to CDN use cases too (and there is a mailing list for possible
ALTO extensions which may or may not need rechartering :) ).

Rich

>
> =A0-- Sebastian
> _______________________________________________
> alto mailing list
> alto@ietf.org
> https://www.ietf.org/mailman/listinfo/alto

From ben@niven-jenkins.co.uk  Wed Feb  1 12:18:50 2012
Return-Path: <ben@niven-jenkins.co.uk>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 20F1611E80E0 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 12:18:50 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.042
X-Spam-Level: 
X-Spam-Status: No, score=-103.042 tagged_above=-999 required=5 tests=[AWL=-0.043, BAYES_00=-2.599, J_CHICKENPOX_51=0.6, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EmXW92YhrWXb for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 12:18:46 -0800 (PST)
Received: from mailex.mailcore.me (mailex.mailcore.me [94.136.40.62]) by ietfa.amsl.com (Postfix) with ESMTP id CFE6E11E80DE for <alto@ietf.org>; Wed,  1 Feb 2012 12:18:45 -0800 (PST)
Received: from cpc10-cmbg15-2-0-cust121.5-4.cable.virginmedia.com ([86.30.246.122] helo=[192.168.0.3]) by mail11.atlas.pipex.net with esmtpa (Exim 4.71) (envelope-from <ben@niven-jenkins.co.uk>) id 1Rsgdg-0000ym-Mk; Wed, 01 Feb 2012 20:18:45 +0000
Mime-Version: 1.0 (Apple Message framework v1084)
Content-Type: text/plain; charset=us-ascii
From: Ben Niven-Jenkins <ben@niven-jenkins.co.uk>
In-Reply-To: <4F2803E7.1090000@bell-labs.com>
Date: Wed, 1 Feb 2012 20:18:42 +0000
Content-Transfer-Encoding: quoted-printable
Message-Id: <C4BDA6FB-21EF-404E-A186-DEB9752D399F@niven-jenkins.co.uk>
References: <A35E0DC7-C133-480C-9967-EC62CAE9BCA0@niven-jenkins.co.uk> <4F272268.4070804@bell-labs.com> <6A051A1C-3C3E-43C7-888C-AB1E98CB2A86@niven-jenkins.co.uk> <489EF99B-832D-4B52-B8ED-5F1612F8BFE2@cisco.com> <4F2803E7.1090000@bell-labs.com>
To: "Vijay K. Gurbani" <vkg@bell-labs.com>
X-Mailer: Apple Mail (2.1084)
X-Mailcore-Auth: 9600544
X-Mailcore-Domain: 172912
Cc: alto <alto@ietf.org>
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 20:18:50 -0000

Vijay,

On 31 Jan 2012, at 15:08, Vijay K. Gurbani wrote:

> On 01/31/2012 05:14 AM, stefano previdi wrote:
> ben> This might be a case where we'd need a PID property to =
distinguish
> ben> "transit" PIDs from "End Host" PIDs.
> ben>
> stefano> sounds pretty reasonable to me.
>=20
> Ben: I suspect that your question had a foundation of some sort
> and was not simply a fishing expedition ;-)

The question does have a foundation :-) Partly I'm trying to understand =
precisely what I can & can not infer from an ALTO map and Bill & Rich's =
follow-up confirm a lot of that for me. We probably want to add some =
text to the protocol spec to clarify that because I suspect if I'm =
asking the question then someone else looking to use ALTO for a purpose =
that wasn't originally in scope and who hasn't been involved with ALTO =
from day 1 might have similar questions.

At the same time I'm trying to think through the applicability of ALTO =
to a number of CDN-related scenarios.

One scenario is whether I can get enough "resolution" out of an ALTO map =
to represent the network topology (as it pertains to a CDN) =
sufficiently. If I need to augment what I can get from ALTO with raw =
IGP/BGP routing to get a sufficient (for the CDN) topology then it =
negates a good chunk of the value provided by ALTO IMO. It's worth =
clarifying here that by "sufficient topology for a CDN" I don't mean =
being able to translate a network map back into the raw network =
topology, I only need enough topology to overlay a CDN topology on top =
of. After thinking a bit more about this I think ALTO as-is probably =
provides sufficient information for the purpose although not in the most =
efficient form (but that's not showstopper).

The other scenarios relate to using ALTO maps for more than just CDN =
cache selection and in those scenarios I'm thinking that ALTO is still =
probably OK and I can achieve a lot of what I need without needing ALTO =
to expose me a graph, provided there are some common constraints applied =
by the ALTO server (like not renaming PIDs) but I need to give those =
scenarios some more thought. I might have a couple of ideas for =
extensions once I'm done (like enabling entries in the IRD to be given a =
name/ID) but that's a separate topic.=20
=20
> Let me ask you this: is the PID property that you would like to put
> in useful for CDN-usage in ALTO?  If so, and since others
> (Stefano) appear to agree, then it is worth a larger discussion on
> the need of such a property to support emerging usages of ALTO.

I'm starting to conclude I don't need the PID property to achieve what I =
want but I haven't 100% convinced myself of that yet. I think the idea =
of PID properties is interesting though so if there's a discussion on =
them I'll get involved. I might have some other use cases but I haven;t =
really given them any thought yet.

Ben

>=20
> Thanks,
>=20
> - vijay
> --=20
> Vijay K. Gurbani, Bell Laboratories, Alcatel-Lucent
> 1960 Lucent Lane, Rm. 9C-533, Naperville, Illinois 60566 (USA)
> Email: vkg@{bell-labs.com,acm.org} / vijay.gurbani@alcatel-lucent.com
> Web:   http://ect.bell-labs.com/who/vkg/


From sebi@gw01.ehlo.wurstkaes.de  Wed Feb  1 14:49:25 2012
Return-Path: <sebi@gw01.ehlo.wurstkaes.de>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 422A811E80A1 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 14:49:25 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.841
X-Spam-Level: 
X-Spam-Status: No, score=-1.841 tagged_above=-999 required=5 tests=[AWL=0.408,  BAYES_00=-2.599, HELO_EQ_DE=0.35]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ABJM2NMXjrFj for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 14:49:24 -0800 (PST)
Received: from gw01.ehlo.wurstkaes.de (gw01.ehlo.wurstkaes.de [188.246.4.151]) by ietfa.amsl.com (Postfix) with ESMTP id 7FB4211E8103 for <alto@ietf.org>; Wed,  1 Feb 2012 14:49:24 -0800 (PST)
Received: from sebi by gw01.ehlo.wurstkaes.de with local (Exim 4.72) (envelope-from <sebi@gw01.ehlo.wurstkaes.de>) id 1Rsiz7-0007s6-3C; Wed, 01 Feb 2012 23:49:01 +0100
Date: Wed, 1 Feb 2012 23:49:01 +0100
From: Sebastian Kiesel <ietf-alto@skiesel.de>
To: stefano previdi <sprevidi@cisco.com>
Message-ID: <20120201224900.GA29210@gw01.ehlo.wurstkaes.de>
References: <mailman.4534.1328093779.3200.alto@ietf.org> <CB4EBC77.A75E%w.roome@alcatel-lucent.com> <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de> <73FCC4FE-5CEE-43A9-B8F1-8BD45D5DFFDB@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <73FCC4FE-5CEE-43A9-B8F1-8BD45D5DFFDB@cisco.com>
Accept-Languages: en, de
Organization: my personal mail account
User-Agent: Mutt/1.5.20 (2009-06-14)
Cc: Bill Roome <w.roome@alcatel-lucent.com>, alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Feb 2012 22:49:25 -0000

Stefano, all,

On Wed, Feb 01, 2012 at 06:04:23PM +0100, stefano previdi wrote:
> > Btw: which application needs to know the topology?
> 
> E.g.: a CDN may want to operate autonomously with maps rather than 
> using ECS. 

Downloading a map and operating autonomously with it is one thing,
trying to derive missing fields in the map/matrix by some kind
of triangulation is another thing.

> Not that I'm advocating this but it all depends on where 
> do you want to place the complexity of topology hint: inside ALTO 
> server (ECS like of approach) or inside the application (Maps that 
> allows you to compute paths/trees).

Maybe part of the confusion is that we need a clear definition what
end-to-end is. I'd say that ALTO should always give the end-to-end
costs for sending IP packets between two endpoints using "normal"
IP-layer forwarding.  Based on that knowledge one can of course
build efficient paths/trees that involve proxies, media relays
or the like (i.e., this higher-layer path is a concatenation of
several IP-layer end-to-end paths). Does this make sense?

> > If I remember
> > correctly once upon a time our primary goal was assisting peer selection
> > (i.e., I know that I can get the desired resource form IP address A, B,
> > or C, so please ALTO service tell me which address I should give a
> > connect()-try first). To that end, the 1xN vector would be enough.
> 
> The CDN is not substantially different in its logic.

this is my feeling as well, that's why I was asking.


   Thanks, 
   Sebastian

From sprevidi@cisco.com  Wed Feb  1 22:45:48 2012
Return-Path: <sprevidi@cisco.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1BA4921F8AD9 for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 22:45:48 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.449
X-Spam-Level: 
X-Spam-Status: No, score=-102.449 tagged_above=-999 required=5 tests=[AWL=0.150, BAYES_00=-2.599, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id IADXjthhRrMp for <alto@ietfa.amsl.com>; Wed,  1 Feb 2012 22:45:47 -0800 (PST)
Received: from av-tac-bru.cisco.com (weird-brew.cisco.com [144.254.15.118]) by ietfa.amsl.com (Postfix) with ESMTP id DEEAE21F8ACD for <alto@ietf.org>; Wed,  1 Feb 2012 22:45:46 -0800 (PST)
X-TACSUNS: Virus Scanned
Received: from stew-brew.cisco.com (localhost.cisco.com [127.0.0.1]) by av-tac-bru.cisco.com (8.13.8+Sun/8.13.8) with ESMTP id q126jjiB000641 for <alto@ietf.org>; Thu, 2 Feb 2012 07:45:45 +0100 (CET)
Received: from ams3-vpn-dhcp5561.cisco.com (ams3-vpn-dhcp5561.cisco.com [10.61.85.184]) by stew-brew.cisco.com (8.13.8+Sun/8.13.8) with ESMTP id q126jh3c003737; Thu, 2 Feb 2012 07:45:43 +0100 (CET)
Mime-Version: 1.0 (Apple Message framework v1251.1)
Content-Type: text/plain; charset=us-ascii
From: stefano previdi <sprevidi@cisco.com>
In-Reply-To: <20120201224900.GA29210@gw01.ehlo.wurstkaes.de>
Date: Thu, 2 Feb 2012 07:45:45 +0100
Content-Transfer-Encoding: 7bit
Message-Id: <C020BC75-5EE7-4194-8BB9-398E8FCF969C@cisco.com>
References: <mailman.4534.1328093779.3200.alto@ietf.org> <CB4EBC77.A75E%w.roome@alcatel-lucent.com> <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de> <73FCC4FE-5CEE-43A9-B8F1-8BD45D5DFFDB@cisco.com> <20120201224900.GA29210@gw01.ehlo.wurstkaes.de>
To: Sebastian Kiesel <ietf-alto@skiesel.de>
X-Mailer: Apple Mail (2.1251.1)
Cc: Bill Roome <w.roome@alcatel-lucent.com>, alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 02 Feb 2012 06:45:48 -0000

On Feb 1, 2012, at 11:49 PM, Sebastian Kiesel wrote:

> Stefano, all,
> 
> On Wed, Feb 01, 2012 at 06:04:23PM +0100, stefano previdi wrote:
>>> Btw: which application needs to know the topology?
>> 
>> E.g.: a CDN may want to operate autonomously with maps rather than 
>> using ECS. 
> 
> Downloading a map and operating autonomously with it is one thing,
> trying to derive missing fields in the map/matrix by some kind
> of triangulation is another thing.
> 
>> Not that I'm advocating this but it all depends on where 
>> do you want to place the complexity of topology hint: inside ALTO 
>> server (ECS like of approach) or inside the application (Maps that 
>> allows you to compute paths/trees).
> 
> Maybe part of the confusion is that we need a clear definition what
> end-to-end is. I'd say that ALTO should always give the end-to-end
> costs for sending IP packets between two endpoints using "normal"
> IP-layer forwarding.  Based on that knowledge one can of course
> build efficient paths/trees that involve proxies, media relays
> or the like (i.e., this higher-layer path is a concatenation of
> several IP-layer end-to-end paths). Does this make sense?


why do you believe alto should only deliver end-to-end costs ?
In a slightly different context it could even deliver a path.


>>> If I remember
>>> correctly once upon a time our primary goal was assisting peer selection
>>> (i.e., I know that I can get the desired resource form IP address A, B,
>>> or C, so please ALTO service tell me which address I should give a
>>> connect()-try first). To that end, the 1xN vector would be enough.
>> 
>> The CDN is not substantially different in its logic.
> 
> this is my feeling as well, that's why I was asking.


there are case other than p2p or cdn. Think about cloud 
networking, data center resources advertisement and distribution, 
DNS optimizations, Demand Engineering, ...

If we want to think about extending alto to other (non p2p) case, 
these are interesting examples.


s.


> 
> 
>   Thanks, 
>   Sebastian
> 


From sebi@gw01.ehlo.wurstkaes.de  Thu Feb  2 04:05:18 2012
Return-Path: <sebi@gw01.ehlo.wurstkaes.de>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8A9B321F8A87 for <alto@ietfa.amsl.com>; Thu,  2 Feb 2012 04:05:18 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.943
X-Spam-Level: 
X-Spam-Status: No, score=-1.943 tagged_above=-999 required=5 tests=[AWL=0.306,  BAYES_00=-2.599, HELO_EQ_DE=0.35]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mebDtfuMRK6u for <alto@ietfa.amsl.com>; Thu,  2 Feb 2012 04:05:18 -0800 (PST)
Received: from gw01.ehlo.wurstkaes.de (gw01.ehlo.wurstkaes.de [188.246.4.151]) by ietfa.amsl.com (Postfix) with ESMTP id EC07C21F8A48 for <alto@ietf.org>; Thu,  2 Feb 2012 04:05:17 -0800 (PST)
Received: from sebi by gw01.ehlo.wurstkaes.de with local (Exim 4.72) (envelope-from <sebi@gw01.ehlo.wurstkaes.de>) id 1RsvPa-0003m1-IL; Thu, 02 Feb 2012 13:05:10 +0100
Date: Thu, 2 Feb 2012 13:05:10 +0100
From: Sebastian Kiesel <ietf-alto@skiesel.de>
To: stefano previdi <sprevidi@cisco.com>
Message-ID: <20120202120510.GB29210@gw01.ehlo.wurstkaes.de>
References: <mailman.4534.1328093779.3200.alto@ietf.org> <CB4EBC77.A75E%w.roome@alcatel-lucent.com> <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de> <73FCC4FE-5CEE-43A9-B8F1-8BD45D5DFFDB@cisco.com> <20120201224900.GA29210@gw01.ehlo.wurstkaes.de> <C020BC75-5EE7-4194-8BB9-398E8FCF969C@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <C020BC75-5EE7-4194-8BB9-398E8FCF969C@cisco.com>
Accept-Languages: en, de
Organization: my personal mail account
User-Agent: Mutt/1.5.20 (2009-06-14)
Cc: alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 02 Feb 2012 12:05:18 -0000

Stefano,

On Thu, Feb 02, 2012 at 07:45:45AM +0100, stefano previdi wrote:
> On Feb 1, 2012, at 11:49 PM, Sebastian Kiesel wrote:
> >> Not that I'm advocating this but it all depends on where 
> >> do you want to place the complexity of topology hint: inside ALTO 
> >> server (ECS like of approach) or inside the application (Maps that 
> >> allows you to compute paths/trees).
> > 
> > Maybe part of the confusion is that we need a clear definition what
> > end-to-end is. I'd say that ALTO should always give the end-to-end
> > costs for sending IP packets between two endpoints using "normal"
> > IP-layer forwarding.  Based on that knowledge one can of course
> > build efficient paths/trees that involve proxies, media relays
> > or the like (i.e., this higher-layer path is a concatenation of
> > several IP-layer end-to-end paths). Does this make sense?
> 
> 
> why do you believe alto should only deliver end-to-end costs ?
> In a slightly different context it could even deliver a path.

what exactly do you mean by "deliver a path"?

And would that require that a host and/or the ALTO server can influence
how packets travel through the network (source routing, take actively
part in OSPF/BGP/... , etc)?


> there are case other than p2p or cdn. Think about cloud 
> networking, data center resources advertisement and distribution, 
> DNS optimizations, Demand Engineering, ...
> 
> If we want to think about extending alto to other (non p2p) case, 
> these are interesting examples.

sure. extending the scope of ALTO makes sense.

but what is different to the P2P use case? One difference is that
- as Rich pointed out - due to other trust relationships we may be able
to expose more information. OK, but how do we use it?  Is this more than
just shifting complexity?

Thanks,
Sebastian

From sprevidi@cisco.com  Thu Feb  2 05:55:06 2012
Return-Path: <sprevidi@cisco.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D04A821F85EC for <alto@ietfa.amsl.com>; Thu,  2 Feb 2012 05:55:06 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.47
X-Spam-Level: 
X-Spam-Status: No, score=-102.47 tagged_above=-999 required=5 tests=[AWL=0.129, BAYES_00=-2.599, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fiH4bCzQW7dd for <alto@ietfa.amsl.com>; Thu,  2 Feb 2012 05:55:06 -0800 (PST)
Received: from av-tac-bru.cisco.com (weird-brew.cisco.com [144.254.15.118]) by ietfa.amsl.com (Postfix) with ESMTP id 0D8E221F85E6 for <alto@ietf.org>; Thu,  2 Feb 2012 05:55:05 -0800 (PST)
X-TACSUNS: Virus Scanned
Received: from stew-brew.cisco.com (localhost.cisco.com [127.0.0.1]) by av-tac-bru.cisco.com (8.13.8+Sun/8.13.8) with ESMTP id q12Dt44t023798 for <alto@ietf.org>; Thu, 2 Feb 2012 14:55:05 +0100 (CET)
Received: from ams3-vpn-dhcp4341.cisco.com (ams3-vpn-dhcp4341.cisco.com [10.61.80.244]) by stew-brew.cisco.com (8.13.8+Sun/8.13.8) with ESMTP id q12Dt47h023590; Thu, 2 Feb 2012 14:55:04 +0100 (CET)
Mime-Version: 1.0 (Apple Message framework v1251.1)
Content-Type: text/plain; charset=us-ascii
From: stefano previdi <sprevidi@cisco.com>
In-Reply-To: <20120202120510.GB29210@gw01.ehlo.wurstkaes.de>
Date: Thu, 2 Feb 2012 14:55:05 +0100
Content-Transfer-Encoding: 7bit
Message-Id: <5AF66283-0527-48F3-9A26-46482E17EBE8@cisco.com>
References: <mailman.4534.1328093779.3200.alto@ietf.org> <CB4EBC77.A75E%w.roome@alcatel-lucent.com> <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de> <73FCC4FE-5CEE-43A9-B8F1-8BD45D5DFFDB@cisco.com> <20120201224900.GA29210@gw01.ehlo.wurstkaes.de> <C020BC75-5EE7-4194-8BB9-398E8FCF969C@cisco.com> <20120202120510.GB29210@gw01.ehlo.wurstkaes.de>
To: Sebastian Kiesel <ietf-alto@skiesel.de>
X-Mailer: Apple Mail (2.1251.1)
Cc: alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 02 Feb 2012 13:55:06 -0000

On Feb 2, 2012, at 1:05 PM, Sebastian Kiesel wrote:

> Stefano,
> 
> On Thu, Feb 02, 2012 at 07:45:45AM +0100, stefano previdi wrote:
>> On Feb 1, 2012, at 11:49 PM, Sebastian Kiesel wrote:
>>>> Not that I'm advocating this but it all depends on where 
>>>> do you want to place the complexity of topology hint: inside ALTO 
>>>> server (ECS like of approach) or inside the application (Maps that 
>>>> allows you to compute paths/trees).
>>> 
>>> Maybe part of the confusion is that we need a clear definition what
>>> end-to-end is. I'd say that ALTO should always give the end-to-end
>>> costs for sending IP packets between two endpoints using "normal"
>>> IP-layer forwarding.  Based on that knowledge one can of course
>>> build efficient paths/trees that involve proxies, media relays
>>> or the like (i.e., this higher-layer path is a concatenation of
>>> several IP-layer end-to-end paths). Does this make sense?
>> 
>> 
>> why do you believe alto should only deliver end-to-end costs ?
>> In a slightly different context it could even deliver a path.
> 
> what exactly do you mean by "deliver a path"?
> 
> And would that require that a host and/or the ALTO server can influence
> how packets travel through the network (source routing, take actively
> part in OSPF/BGP/... , etc)?


well, I don't go that far. However, an application may ask for network 
guidance and get some preferences (rankings) computed by an alto server 
that goes along with some form of path description associated to that 
preference.


>> there are case other than p2p or cdn. Think about cloud 
>> networking, data center resources advertisement and distribution, 
>> DNS optimizations, Demand Engineering, ...
>> 
>> If we want to think about extending alto to other (non p2p) case, 
>> these are interesting examples.
> 
> sure. extending the scope of ALTO makes sense.
> 
> but what is different to the P2P use case?


p2p and cdn have some similarities but there are a few other 
substantially different cases.

s.


> One difference is that
> - as Rich pointed out - due to other trust relationships we may be able
> to expose more information. OK, but how do we use it?  Is this more than
> just shifting complexity?
> 
> Thanks,
> Sebastian
> 


From sebi@gw01.ehlo.wurstkaes.de  Thu Feb  2 08:36:01 2012
Return-Path: <sebi@gw01.ehlo.wurstkaes.de>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1C32421F8615 for <alto@ietfa.amsl.com>; Thu,  2 Feb 2012 08:36:01 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.004
X-Spam-Level: 
X-Spam-Status: No, score=-2.004 tagged_above=-999 required=5 tests=[AWL=0.245,  BAYES_00=-2.599, HELO_EQ_DE=0.35]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id jGMDc-hyysZM for <alto@ietfa.amsl.com>; Thu,  2 Feb 2012 08:36:00 -0800 (PST)
Received: from gw01.ehlo.wurstkaes.de (gw01.ehlo.wurstkaes.de [188.246.4.151]) by ietfa.amsl.com (Postfix) with ESMTP id 7820421F8613 for <alto@ietf.org>; Thu,  2 Feb 2012 08:36:00 -0800 (PST)
Received: from sebi by gw01.ehlo.wurstkaes.de with local (Exim 4.72) (envelope-from <sebi@gw01.ehlo.wurstkaes.de>) id 1RszdZ-00052a-J7; Thu, 02 Feb 2012 17:35:53 +0100
Date: Thu, 2 Feb 2012 17:35:53 +0100
From: Sebastian Kiesel <ietf-alto@skiesel.de>
To: stefano previdi <sprevidi@cisco.com>
Message-ID: <20120202163553.GD29210@gw01.ehlo.wurstkaes.de>
References: <mailman.4534.1328093779.3200.alto@ietf.org> <CB4EBC77.A75E%w.roome@alcatel-lucent.com> <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de> <73FCC4FE-5CEE-43A9-B8F1-8BD45D5DFFDB@cisco.com> <20120201224900.GA29210@gw01.ehlo.wurstkaes.de> <C020BC75-5EE7-4194-8BB9-398E8FCF969C@cisco.com> <20120202120510.GB29210@gw01.ehlo.wurstkaes.de> <5AF66283-0527-48F3-9A26-46482E17EBE8@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <5AF66283-0527-48F3-9A26-46482E17EBE8@cisco.com>
Accept-Languages: en, de
Organization: my personal mail account
User-Agent: Mutt/1.5.20 (2009-06-14)
Cc: alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 02 Feb 2012 16:36:01 -0000

On Thu, Feb 02, 2012 at 02:55:05PM +0100, stefano previdi wrote:
> On Feb 2, 2012, at 1:05 PM, Sebastian Kiesel wrote:
> > On Thu, Feb 02, 2012 at 07:45:45AM +0100, stefano previdi wrote:
> >> On Feb 1, 2012, at 11:49 PM, Sebastian Kiesel wrote:
> >>>> Not that I'm advocating this but it all depends on where 
> >>>> do you want to place the complexity of topology hint: inside ALTO 
> >>>> server (ECS like of approach) or inside the application (Maps that 
> >>>> allows you to compute paths/trees).
> >>> 
> >>> Maybe part of the confusion is that we need a clear definition what
> >>> end-to-end is. I'd say that ALTO should always give the end-to-end
> >>> costs for sending IP packets between two endpoints using "normal"
> >>> IP-layer forwarding.  Based on that knowledge one can of course
> >>> build efficient paths/trees that involve proxies, media relays
> >>> or the like (i.e., this higher-layer path is a concatenation of
> >>> several IP-layer end-to-end paths). Does this make sense?
> >> 
> >> 
> >> why do you believe alto should only deliver end-to-end costs ?
> >> In a slightly different context it could even deliver a path.
> > 
> > what exactly do you mean by "deliver a path"?
> > 
> > And would that require that a host and/or the ALTO server can influence
> > how packets travel through the network (source routing, take actively
> > part in OSPF/BGP/... , etc)?
> 
> 
> well, I don't go that far. However, an application may ask for network 
> guidance and get some preferences (rankings) computed by an alto server 
> that goes along with some form of path description associated to that 
> preference.

and what would the application do with that additional information?

if it is used for re-calculating ALTO's preferences, why can't that
be done in the ALTO server in the first place?

Thanks,
Sebastian

From sprevidi@cisco.com  Thu Feb  2 08:39:03 2012
Return-Path: <sprevidi@cisco.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5F0E621F861B for <alto@ietfa.amsl.com>; Thu,  2 Feb 2012 08:39:03 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.487
X-Spam-Level: 
X-Spam-Status: No, score=-102.487 tagged_above=-999 required=5 tests=[AWL=0.113, BAYES_00=-2.599, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NrZHTkUKWT1B for <alto@ietfa.amsl.com>; Thu,  2 Feb 2012 08:39:02 -0800 (PST)
Received: from av-tac-bru.cisco.com (weird-brew.cisco.com [144.254.15.118]) by ietfa.amsl.com (Postfix) with ESMTP id 96CFA21F853D for <alto@ietf.org>; Thu,  2 Feb 2012 08:39:02 -0800 (PST)
X-TACSUNS: Virus Scanned
Received: from stew-brew.cisco.com (localhost.cisco.com [127.0.0.1]) by av-tac-bru.cisco.com (8.13.8+Sun/8.13.8) with ESMTP id q12Gd1gW013259 for <alto@ietf.org>; Thu, 2 Feb 2012 17:39:01 +0100 (CET)
Received: from ams3-vpn-dhcp4341.cisco.com (ams3-vpn-dhcp4341.cisco.com [10.61.80.244]) by stew-brew.cisco.com (8.13.8+Sun/8.13.8) with ESMTP id q12GcvBJ028110; Thu, 2 Feb 2012 17:38:57 +0100 (CET)
Mime-Version: 1.0 (Apple Message framework v1251.1)
Content-Type: text/plain; charset=us-ascii
From: stefano previdi <sprevidi@cisco.com>
In-Reply-To: <20120202163553.GD29210@gw01.ehlo.wurstkaes.de>
Date: Thu, 2 Feb 2012 17:38:57 +0100
Content-Transfer-Encoding: 7bit
Message-Id: <ADAACD74-78A9-4012-B35E-82D105446F31@cisco.com>
References: <mailman.4534.1328093779.3200.alto@ietf.org> <CB4EBC77.A75E%w.roome@alcatel-lucent.com> <20120201165032.GJ3017@gw01.ehlo.wurstkaes.de> <73FCC4FE-5CEE-43A9-B8F1-8BD45D5DFFDB@cisco.com> <20120201224900.GA29210@gw01.ehlo.wurstkaes.de> <C020BC75-5EE7-4194-8BB9-398E8FCF969C@cisco.com> <20120202120510.GB29210@gw01.ehlo.wurstkaes.de> <5AF66283-0527-48F3-9A26-46482E17EBE8@cisco.com> <20120202163553.GD29210@gw01.ehlo.wurstkaes.de>
To: Sebastian Kiesel <ietf-alto@skiesel.de>
X-Mailer: Apple Mail (2.1251.1)
Cc: alto@ietf.org
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 02 Feb 2012 16:39:03 -0000

On Feb 2, 2012, at 5:35 PM, Sebastian Kiesel wrote:

> On Thu, Feb 02, 2012 at 02:55:05PM +0100, stefano previdi wrote:
>> On Feb 2, 2012, at 1:05 PM, Sebastian Kiesel wrote:
>>> On Thu, Feb 02, 2012 at 07:45:45AM +0100, stefano previdi wrote:
>>>> On Feb 1, 2012, at 11:49 PM, Sebastian Kiesel wrote:
>>>>>> Not that I'm advocating this but it all depends on where 
>>>>>> do you want to place the complexity of topology hint: inside ALTO 
>>>>>> server (ECS like of approach) or inside the application (Maps that 
>>>>>> allows you to compute paths/trees).
>>>>> 
>>>>> Maybe part of the confusion is that we need a clear definition what
>>>>> end-to-end is. I'd say that ALTO should always give the end-to-end
>>>>> costs for sending IP packets between two endpoints using "normal"
>>>>> IP-layer forwarding.  Based on that knowledge one can of course
>>>>> build efficient paths/trees that involve proxies, media relays
>>>>> or the like (i.e., this higher-layer path is a concatenation of
>>>>> several IP-layer end-to-end paths). Does this make sense?
>>>> 
>>>> 
>>>> why do you believe alto should only deliver end-to-end costs ?
>>>> In a slightly different context it could even deliver a path.
>>> 
>>> what exactly do you mean by "deliver a path"?
>>> 
>>> And would that require that a host and/or the ALTO server can influence
>>> how packets travel through the network (source routing, take actively
>>> part in OSPF/BGP/... , etc)?
>> 
>> 
>> well, I don't go that far. However, an application may ask for network 
>> guidance and get some preferences (rankings) computed by an alto server 
>> that goes along with some form of path description associated to that 
>> preference.
> 
> and what would the application do with that additional information?


the application may make a more accurate choice between two 
endpoint if it known also the characteristics of the paths 
to these endpoints.

s.


> 
> if it is used for re-calculating ALTO's preferences, why can't that
> be done in the ALTO server in the first place?
> 
> Thanks,
> Sebastian
> 


From jmedved@cisco.com  Thu Feb  2 23:22:03 2012
Return-Path: <jmedved@cisco.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7F92721F8576 for <alto@ietfa.amsl.com>; Thu,  2 Feb 2012 23:22:03 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.149
X-Spam-Level: 
X-Spam-Status: No, score=-6.149 tagged_above=-999 required=5 tests=[AWL=-0.150, BAYES_00=-2.599, J_CHICKENPOX_51=0.6, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id BvHpFQi8o1de for <alto@ietfa.amsl.com>; Thu,  2 Feb 2012 23:22:02 -0800 (PST)
Received: from mtv-iport-1.cisco.com (mtv-iport-1.cisco.com [173.36.130.12]) by ietfa.amsl.com (Postfix) with ESMTP id 9AB3321F84E2 for <alto@ietf.org>; Thu,  2 Feb 2012 23:22:02 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=jmedved@cisco.com; l=4270; q=dns/txt; s=iport; t=1328253722; x=1329463322; h=subject:mime-version:from:in-reply-to:date:cc: content-transfer-encoding:message-id:references:to; bh=4sWgduKKbZXEHBI7aZotVK3+G0IxMn+gTuIL7M4RmEU=; b=PiVc4UJcQZiiPbD8TPyfdLIQAd/YzAAcZS4fP6E2dzkxT1vjFU4G5UMx Znx3IUQ4oWli+lWQEbpjduavQqp58Um4/5lF40LUli+Vl/vDa750KdYsC jbhD1zh+BHPQ6C3D1QFz7XdCnkxFCQc9SG5sW9ovbkz3X7UaFX/0JANue A=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: Av4EAACKK0+rRDoG/2dsb2JhbABDrxiBBYFyAQEBAwEBAQEPASc0CwULCxguJzAGEyKHWgmZagGeX4tGBwICCQUMBhMBCAUDAwkFDAKDCQUYAgsCBYETglpjBIhCjGGSdw
X-IronPort-AV: E=Sophos;i="4.73,350,1325462400"; d="scan'208";a="26835252"
Received: from mtv-core-1.cisco.com ([171.68.58.6]) by mtv-iport-1.cisco.com with ESMTP; 03 Feb 2012 07:22:02 +0000
Received: from sjc-vpn3-708.cisco.com (sjc-vpn3-708.cisco.com [10.21.66.196]) by mtv-core-1.cisco.com (8.14.3/8.14.3) with ESMTP id q137M295018573; Fri, 3 Feb 2012 07:22:02 GMT
Mime-Version: 1.0 (Apple Message framework v1251.1)
Content-Type: text/plain; charset=us-ascii
From: Jan Medved <jmedved@cisco.com>
In-Reply-To: <C4BDA6FB-21EF-404E-A186-DEB9752D399F@niven-jenkins.co.uk>
Date: Thu, 2 Feb 2012 22:55:06 -0800
Content-Transfer-Encoding: quoted-printable
Message-Id: <8800B6E9-E618-4642-ADD9-38929600AA9F@cisco.com>
References: <A35E0DC7-C133-480C-9967-EC62CAE9BCA0@niven-jenkins.co.uk> <4F272268.4070804@bell-labs.com> <6A051A1C-3C3E-43C7-888C-AB1E98CB2A86@niven-jenkins.co.uk> <489EF99B-832D-4B52-B8ED-5F1612F8BFE2@cisco.com> <4F2803E7.1090000@bell-labs.com> <C4BDA6FB-21EF-404E-A186-DEB9752D399F@niven-jenkins.co.uk>
To: Ben Niven-Jenkins <ben@niven-jenkins.co.uk>
X-Mailer: Apple Mail (2.1251.1)
Cc: alto <alto@ietf.org>
Subject: Re: [alto] Are ALTO costs always End to End?
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 03 Feb 2012 07:22:03 -0000

On Feb 1, 2012, at 12:18 PM, Ben Niven-Jenkins wrote:

> Vijay,
>=20
> On 31 Jan 2012, at 15:08, Vijay K. Gurbani wrote:
>=20
>> On 01/31/2012 05:14 AM, stefano previdi wrote:
>> ben> This might be a case where we'd need a PID property to =
distinguish
>> ben> "transit" PIDs from "End Host" PIDs.
>> ben>
>> stefano> sounds pretty reasonable to me.
>>=20
>> Ben: I suspect that your question had a foundation of some sort
>> and was not simply a fishing expedition ;-)
>=20
> The question does have a foundation :-) Partly I'm trying to =
understand precisely what I can & can not infer from an ALTO map and =
Bill & Rich's follow-up confirm a lot of that for me. We probably want =
to add some text to the protocol spec to clarify that because I suspect =
if I'm asking the question then someone else looking to use ALTO for a =
purpose that wasn't originally in scope and who hasn't been involved =
with ALTO from day 1 might have similar questions.
>=20
> At the same time I'm trying to think through the applicability of ALTO =
to a number of CDN-related scenarios.
>=20
> One scenario is whether I can get enough "resolution" out of an ALTO =
map to represent the network topology (as it pertains to a CDN) =
sufficiently.

Yes you can. ALTO servers providing this functionality have been =
designed, implemented and some of them are in production.

> If I need to augment what I can get from ALTO with raw IGP/BGP routing =
to get a sufficient (for the CDN) topology then it negates a good chunk =
of the value provided by ALTO IMO.

You don't need to augment ALTO with IGP/BGP data. The ALTO data model =
has enough flexibility to convey sufficient topology information from =
IGP/BGP to a CDN. Again, ALTO servers that do this have been designed, =
implemented and some of them are in production. CDN Request Routers that =
utilize topology information from an ALTO Server have been designed, =
implemented and some of them are in production.=20

> It's worth clarifying here that by "sufficient topology for a CDN" I =
don't mean being able to translate a network map back into the raw =
network topology, I only need enough topology to overlay a CDN topology =
on top of. After thinking a bit more about this I think ALTO as-is =
probably provides sufficient information for the purpose although not in =
the most efficient form (but that's not showstopper).
>=20
We are in agreement here. Even the efficiency is good enough, especially =
if you use the Endpoint Cost Interface.

> The other scenarios relate to using ALTO maps for more than just CDN =
cache selection and in those scenarios I'm thinking that ALTO is still =
probably OK and I can achieve a lot of what I need without needing ALTO =
to expose me a graph, provided there are some common constraints applied =
by the ALTO server (like not renaming PIDs) but I need to give those =
scenarios some more thought. I might have a couple of ideas for =
extensions once I'm done (like enabling entries in the IRD to be given a =
name/ID) but that's a separate topic.=20
>=20
>> Let me ask you this: is the PID property that you would like to put
>> in useful for CDN-usage in ALTO?  If so, and since others
>> (Stefano) appear to agree, then it is worth a larger discussion on
>> the need of such a property to support emerging usages of ALTO.
>=20
> I'm starting to conclude I don't need the PID property to achieve what =
I want but I haven't 100% convinced myself of that yet. I think the idea =
of PID properties is interesting though so if there's a discussion on =
them I'll get involved. I might have some other use cases but I haven;t =
really given them any thought yet.
>=20
PID properties (a.k.a PID attributes) are a good idea. "Transit" PIDs =
are not.

> Ben
>=20

/Jan

>>=20
>> Thanks,
>>=20
>> - vijay
>> --=20
>> Vijay K. Gurbani, Bell Laboratories, Alcatel-Lucent
>> 1960 Lucent Lane, Rm. 9C-533, Naperville, Illinois 60566 (USA)
>> Email: vkg@{bell-labs.com,acm.org} / vijay.gurbani@alcatel-lucent.com
>> Web:   http://ect.bell-labs.com/who/vkg/
>=20
> _______________________________________________
> alto mailing list
> alto@ietf.org
> https://www.ietf.org/mailman/listinfo/alto


From lmcm@tid.es  Fri Feb  3 00:23:10 2012
Return-Path: <lmcm@tid.es>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 40BC221F853F for <alto@ietfa.amsl.com>; Fri,  3 Feb 2012 00:23:10 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.87
X-Spam-Level: 
X-Spam-Status: No, score=-1.87 tagged_above=-999 required=5 tests=[AWL=-1.872,  BAYES_50=0.001, HTML_MESSAGE=0.001]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id y87dR8iZM5Bm for <alto@ietfa.amsl.com>; Fri,  3 Feb 2012 00:23:08 -0800 (PST)
Received: from tidos.tid.es (tidos.tid.es [195.235.93.44]) by ietfa.amsl.com (Postfix) with ESMTP id 34E0E21F8535 for <alto@ietf.org>; Fri,  3 Feb 2012 00:23:06 -0800 (PST)
Received: from sbrightmailg01.hi.inet (sbrightmailg01.hi.inet [10.95.64.104]) by tid.hi.inet (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0LYT00HCY5YGHA@tid.hi.inet> for alto@ietf.org; Fri, 03 Feb 2012 09:23:04 +0100 (MET)
Received: from tid (tid.hi.inet [10.95.64.10])	by sbrightmailg01.hi.inet (Symantec Messaging Gateway) with SMTP id 1D.B9.02893.8699B2F4; Fri, 03 Feb 2012 09:23:04 +0100 (CET)
Received: from correo.tid.es (mailhost.hi.inet [10.95.64.100]) by tid.hi.inet (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTPS id <0LYT00HCV5YGHA@tid.hi.inet> for alto@ietf.org; Fri, 03 Feb 2012 09:23:04 +0100 (MET)
Received: from EXCLU2K7.hi.inet ([10.95.67.65]) by htcasmad2.hi.inet ([192.168.0.2]) with mapi; Fri, 03 Feb 2012 09:23:04 +0100
Date: Fri, 03 Feb 2012 09:23:03 +0100
From: LUIS MIGUEL CONTRERAS MURILLO <lmcm@tid.es>
To: "alto@ietf.org" <alto@ietf.org>
Message-id: <B348B152E5F11640B2247E54304E53FC590D5C8B79@EXCLU2K7.hi.inet>
MIME-version: 1.0
Content-type: multipart/alternative; boundary="Boundary_(ID_Gc6q6bJGX2z8M3SCETU77g)"
Content-language: es-ES
Accept-Language: es-ES, en-US
Thread-topic: CfP "International Workshop on Cross-Stratum Optimization for Cloud Computing and Distributed Networked Applications"
Thread-index: AcziTDn40f0zVI5QRP6hzLwejHi7IAAALcvA
acceptlanguage: es-ES, en-US
X-AuditID: 0a5f4068-b7f2d6d000000b4d-28-4f2b9968ecdd
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFprBKsWRmVeSWpSXmKPExsXCFe/ApZsxU9vf4MAzI4uH2/eyOjB6LFny kymAMYrLJiU1J7MstUjfLoEr4/sBo4LpUxgrbuxawtjA2N7I2MXIySEhYCLx9PsbJghbTOLC vfVsXYxcHEICGxglep+fY4dwvjNKNLTMAasSEmhklFi5QBDEZhFQlXjwZBPYJDYBQ4lZOyex gjQICzQxSuw7dwCsQURAWeLDpi9gNq+Ap8SJidsZIWxBiR+T77GA2MwCuRKnO+YxQ9jiEnN+ TWQFsRkFZCVWnj/NCDJURKCdUeLQlfdQQ40kzk4/yQxRJCPxf/leFogfBCSW7DnPDGGLSrx8 /I91AqPwLCT7ZiHZNwvJPghbT+LG1ClsELa2xLKFr6FqdCVm/DvEgiy+gJF9FaNYcVJRZnpG SW5iZk66gaFeRqZeZl5qySZGSNRk7GBcvlPlEKMAB6MSD2/HfS1/IdbEsuLK3EOMkhxMSqK8 q6Zq+wvxJeWnVGYkFmfEF5XmpBYfYpTgYFYS4fVqAcrxpiRWVqUW5cOkZDg4lCR4D80ASgkW paanVqRl5gBTA0yaiYMTpJ0HqP0YSA1vcUFibnFmOkT+FKMqx6O3384zCrHk5eelSonzLgQp EgApyijNg5vzilEc6GBh3rkgWR5gcoOb8ApoOBPQcAYLTZDhJYkIKakGRn6zusNiu7elMpa7 1795YjmFmfVHnOvzO9Xcq1RS3jSLvrKce6c7ZtUMo4atd3dpTz1RdVPPMmiurlr63Ak99yoO 7Fqi7V227EdaUN88r0n2mh9XeysvcNXvfum31a6cYS3H0dcv44M7jmwINln1wXPubaD7d/3d MrWYT/ijqqrOS/f+vom5SizFGYmGWsxFxYkAYT/sgysDAAA=
Subject: [alto] CfP "International Workshop on Cross-Stratum Optimization for Cloud Computing and Distributed Networked Applications"
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 03 Feb 2012 08:23:10 -0000

--Boundary_(ID_Gc6q6bJGX2z8M3SCETU77g)
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: quoted-printable

Dear colleagues,

I forward you the following Call for Papers because IMHO it is aligned with=
 the interests of this community.

I think this could be a good place where disseminate original work part of,=
 or based on, the work being carried here.

Sorry for any inconvenience,

Thanks, best regards,

Luis

[Apologies if you receive multiple copies of this message]

Reminder
Call for Papers for the "International Workshop on Cross-Stratum Optimizati=
on for Cloud Computing and Distributed Networked Applications"
http://cccso.net/ or http://www.cccso.net/

Co-located with the 10th IEEE International Symposium on Parallel and Distr=
ibuted Processing with Applications, ISPA 2012 http://www.arcos.inf.uc3m.es=
/ispa12/index.shtml
July 10-13, 2012

Leganes, Madrid, Spain

Aims and Scope
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
The current lack of interaction between networked applications and the unde=
rlying network during service provisioning can cause inefficiencies in the =
use of network resources and can negatively impact the quality received by =
the final consumers of those applications.
Typical networked applications are offered through Information Technology (=
IT) resources (such as computing and storage facilities) residing in data c=
enters. Data centers then provide the physical and virtual infrastructure i=
n which applications and services are provided. Since the data centers are =
usually distributed geographically around a network, many decisions made in=
 the control and management of application services, such as where to insta=
ntiate another service instance, or which data center out of several is ass=
igned to a new customer, can have a significant impact on the state of the =
network. In the same way, the capabilities and state of the network can hav=
e a major impact on application performance.
Cross-stratum optimization (CSO) is defined as the combined optimization of=
 both the IT data center and the network components of an application that =
aims to provide joint resource optimization, responsiveness to quickly chan=
ging demands from/to application to/from network, enhanced service resilien=
ce using cooperative recovery techniques, and quality of experience assuran=
ce by a better use of existing network and application resources, among oth=
ers.
The CSO involves the overall optimization of application layer (IT) and net=
work resources by envisioning next generation architecture for interactions=
 and exchanges between the two layers to improve service continuity, perfor=
mance guarantees, scalability and manageability. The goal of this workshop =
is to promote the research interest on the optimal integration of applicati=
on and network resources.
This workshop aims to explore the challenges and issues faced by cloud comp=
uting and data center integration with networks. Among the key areas of inv=
estigation to be discussed in the workshop are as follows:
.- Application/network integration architectures and subsystems
.- Use cases, business models and requirements for application/network inte=
gration
.- Control/management issues for application/network integration
.- Network virtualization and its impact for application/network integratio=
n
.- Network-aware application/cloud computing
.- Flexible and scalable networking solutions for distributed Data Centers
.- Joint application/network reliability and security
.- Experimental/trial experience
.- Scalability
.- Joint/shared performance and fault monitoring
.- Multi-domain issues

Important Dates
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
Extended Submission Deadline: 17 February 2012 (*firm* deadline)
Paper Acceptance Notification: 30 March 2012
Camera-ready Paper Submissions: 13 April 2012
Tentative workshop day: 10 July 2012 (to be confirmed)

Venue
=3D=3D=3D=3D=3D=3D
Universidad Carlos III de Madrid, Leganes, Madrid, Spain

Submision guidelines
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
Target papers should describe original and unpublished work. All papers mus=
t be submitted in PDF format. Authors should submit full papers of up to 6 =
pages, following strictly the IEEE Computer Society Proceedings Manuscript =
style (available at http://www.computer.org/portal/web/cscps/formatting), u=
sing two-column, single-space format, with 10-point font size. Figures and =
references must be included in the 6 pages.
The submission process will be done online by using EasyChair submission sy=
stem:
http://www.easychair.org/conferences/?conf=3Dispa2012
choosing this workshop among the various workshops held in conjunction with=
 ISPA-2012
Submissions received after deadline, exceeding length limit, or not followi=
ng the specified format will not be considered.
The proceedings will be published by the IEEE in the same volume as the mai=
n conference and will be made online through the IEEE Xplore.

Technical Program Committee
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D
Richard Alimi (Google, USA)
Greg Bernstein (Grotto Networking, USA)
TaeSang Choi (ETRI, Korea)
Nicola Ciulli (Nextworks, Italy)
Oscar Gonzalez de Dios (Telef=F3nica I+D, Spain)
Volker Hilt (Bell Labs, USA)
Giada Landi (Nextworks, Italy)
Dan Li (Huawei, China)
Vishwas Manral (HP, USA)
Thomas D. Nadeau (CA Technologies, USA)
Kohei Shiomoto (NTT, Japan)
Ning So (Verizon, USA)
Hui Yang (Beijing University Post and Telecommunication (BUPT), China)
Yang Richard Yang (Yale University, USA)

Workshop Organizing Committee
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
Young Lee, Huawei Technologies (leeyoung (at) huawei.com)
Luis M. Contreras, Telef=F3nica I+D (lmcm (at) tid.es)
Andrea Fumagalli, The University of Texas at Dallas (andreaf (at) utdallas.=
edu)

_______________
Luis M. Contreras
Telef=F3nica I+D
lmcm@tid.es<mailto:lmcm@tid.es>


________________________________
Este mensaje se dirige exclusivamente a su destinatario. Puede consultar nu=
estra pol=EDtica de env=EDo y recepci=F3n de correo electr=F3nico en el enl=
ace situado m=E1s abajo.
This message is intended exclusively for its addressee. We only send and re=
ceive email on the basis of the terms set out at
http://www.tid.es/ES/PAGINAS/disclaimer.aspx

--Boundary_(ID_Gc6q6bJGX2z8M3SCETU77g)
Content-type: text/html; charset=iso-8859-1
Content-transfer-encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Diso-8859-=
1">
<meta name=3D"Generator" content=3D"Microsoft Word 14 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0cm;
	margin-bottom:.0001pt;
	font-size:11.0pt;
	font-family:"Calibri","sans-serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EstiloCorreo17
	{mso-style-type:personal;
	font-family:"Calibri","sans-serif";
	color:windowtext;}
span.EstiloCorreo18
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-size:10.0pt;}
@page WordSection1
	{size:612.0pt 792.0pt;
	margin:70.85pt 3.0cm 70.85pt 3.0cm;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal">Dear colleagues, <o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">I forward you the following Call for Papers because =
IMHO it is aligned with the interests of this community.
<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">I think this could be a good place where disseminate=
 original work part of, or based on, the work being carried here.<o:p></o:p=
></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Sorry for any inconvenience,<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Thanks, best regards,<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Luis<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">[Apologies if you receive multiple copies of this me=
ssage]<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Reminder <o:p></o:p></p>
<p class=3D"MsoNormal">Call for Papers for the &quot;International Workshop=
 on Cross-Stratum Optimization for Cloud Computing and Distributed Networke=
d Applications&quot;<o:p></o:p></p>
<p class=3D"MsoNormal"><a href=3D"http://cccso.net/">http://cccso.net/</a> =
or <a href=3D"http://www.cccso.net/">
http://www.cccso.net/</a><o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Co-located with the 10th IEEE International Symposiu=
m on Parallel and Distributed Processing with Applications, ISPA 2012
<a href=3D"http://www.arcos.inf.uc3m.es/ispa12/index.shtml">http://www.arco=
s.inf.uc3m.es/ispa12/index.shtml</a><o:p></o:p></p>
<p class=3D"MsoNormal">July 10-13, 2012<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Leganes, Madrid, Spain<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Aims and Scope<o:p></o:p></p>
<p class=3D"MsoNormal">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<o:p></o:p></=
p>
<p class=3D"MsoNormal">The current lack of interaction between networked ap=
plications and the underlying network during service provisioning can cause=
 inefficiencies in the use of network resources and can negatively impact t=
he quality received by the final consumers
 of those applications.<o:p></o:p></p>
<p class=3D"MsoNormal">Typical networked applications are offered through I=
nformation Technology (IT) resources (such as computing and storage facilit=
ies) residing in data centers. Data centers then provide the physical and v=
irtual infrastructure in which applications
 and services are provided. Since the data centers are usually distributed =
geographically around a network, many decisions made in the control and man=
agement of application services, such as where to instantiate another servi=
ce instance, or which data center
 out of several is assigned to a new customer, can have a significant impac=
t on the state of the network. In the same way, the capabilities and state =
of the network can have a major impact on application performance.<o:p></o:=
p></p>
<p class=3D"MsoNormal">Cross-stratum optimization (CSO) is defined as the c=
ombined optimization of both the IT data center and the network components =
of an application that aims to provide joint resource optimization, respons=
iveness to quickly changing demands
 from/to application to/from network, enhanced service resilience using coo=
perative recovery techniques, and quality of experience assurance by a bett=
er use of existing network and application resources, among others.<o:p></o=
:p></p>
<p class=3D"MsoNormal">The CSO involves the overall optimization of applica=
tion layer (IT) and network resources by envisioning next generation archit=
ecture for interactions and exchanges between the two layers to improve ser=
vice continuity, performance guarantees,
 scalability and manageability. The goal of this workshop is to promote the=
 research interest on the optimal integration of application and network re=
sources.<o:p></o:p></p>
<p class=3D"MsoNormal">This workshop aims to explore the challenges and iss=
ues faced by cloud computing and data center integration with networks. Amo=
ng the key areas of investigation to be discussed in the workshop are as fo=
llows:<o:p></o:p></p>
<p class=3D"MsoNormal">.- Application/network integration architectures and=
 subsystems<o:p></o:p></p>
<p class=3D"MsoNormal">.- Use cases, business models and requirements for a=
pplication/network integration<o:p></o:p></p>
<p class=3D"MsoNormal">.- Control/management issues for application/network=
 integration<o:p></o:p></p>
<p class=3D"MsoNormal">.- Network virtualization and its impact for applica=
tion/network integration<o:p></o:p></p>
<p class=3D"MsoNormal">.- Network-aware application/cloud computing<o:p></o=
:p></p>
<p class=3D"MsoNormal">.- Flexible and scalable networking solutions for di=
stributed Data Centers<o:p></o:p></p>
<p class=3D"MsoNormal">.- Joint application/network reliability and securit=
y<o:p></o:p></p>
<p class=3D"MsoNormal">.- Experimental/trial experience<o:p></o:p></p>
<p class=3D"MsoNormal">.- Scalability<o:p></o:p></p>
<p class=3D"MsoNormal">.- Joint/shared performance and fault monitoring<o:p=
></o:p></p>
<p class=3D"MsoNormal">.- Multi-domain issues<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Important Dates<o:p></o:p></p>
<p class=3D"MsoNormal">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<o:p></=
o:p></p>
<p class=3D"MsoNormal">Extended Submission Deadline: 17 February 2012 (*fir=
m* deadline)<o:p></o:p></p>
<p class=3D"MsoNormal">Paper Acceptance Notification: 30 March 2012 <o:p></=
o:p></p>
<p class=3D"MsoNormal">Camera-ready Paper Submissions: 13 April 2012 <o:p><=
/o:p></p>
<p class=3D"MsoNormal">Tentative workshop day: 10 July 2012 (to be confirme=
d)<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Venue<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"ES">=3D=3D=3D=3D=3D=3D<o:p></o:p></spa=
n></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Universidad Carlos III de Madrid, =
Leganes, Madrid, Spain<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"ES"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal">Submision guidelines<o:p></o:p></p>
<p class=3D"MsoNormal">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D<o:p></o:p></p>
<p class=3D"MsoNormal">Target papers should describe original and unpublish=
ed work. All papers must be submitted in PDF format. Authors should submit =
full papers of up to 6 pages, following strictly the IEEE Computer Society =
Proceedings Manuscript style (available
 at <a href=3D"http://www.computer.org/portal/web/cscps/formatting">http://=
www.computer.org/portal/web/cscps/formatting</a>), using two-column, single=
-space format, with 10-point font size. Figures and references must be incl=
uded in the 6 pages.
<o:p></o:p></p>
<p class=3D"MsoNormal">The submission process will be done online by using =
EasyChair submission system:<o:p></o:p></p>
<p class=3D"MsoNormal"><a href=3D"http://www.easychair.org/conferences/?con=
f=3Dispa2012">http://www.easychair.org/conferences/?conf=3Dispa2012</a><o:p=
></o:p></p>
<p class=3D"MsoNormal">choosing this workshop among the various workshops h=
eld in conjunction with ISPA-2012
<o:p></o:p></p>
<p class=3D"MsoNormal">Submissions received after deadline, exceeding lengt=
h limit, or not following the specified format will not be considered.<o:p>=
</o:p></p>
<p class=3D"MsoNormal">The proceedings will be published by the IEEE in the=
 same volume as the main conference and will be made online through the IEE=
E Xplore.<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Technical Program Committee<o:p></o:p></p>
<p class=3D"MsoNormal">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<o:p></o:p></p>
<p class=3D"MsoNormal">Richard Alimi (Google, USA)<o:p></o:p></p>
<p class=3D"MsoNormal">Greg Bernstein (Grotto Networking, USA)<o:p></o:p></=
p>
<p class=3D"MsoNormal"><span lang=3D"ES">TaeSang Choi (ETRI, Korea)<o:p></o=
:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Nicola Ciulli (Nextworks, Italy)<o=
:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Oscar Gonzalez de Dios (Telef=F3ni=
ca I&#43;D, Spain)
<o:p></o:p></span></p>
<p class=3D"MsoNormal">Volker Hilt (Bell Labs, USA) <o:p></o:p></p>
<p class=3D"MsoNormal">Giada Landi (Nextworks, Italy) <o:p></o:p></p>
<p class=3D"MsoNormal">Dan Li (Huawei, China) <o:p></o:p></p>
<p class=3D"MsoNormal">Vishwas Manral (HP, USA)<o:p></o:p></p>
<p class=3D"MsoNormal">Thomas D. Nadeau (CA Technologies, USA) <o:p></o:p><=
/p>
<p class=3D"MsoNormal">Kohei Shiomoto (NTT, Japan) <o:p></o:p></p>
<p class=3D"MsoNormal">Ning So (Verizon, USA) <o:p></o:p></p>
<p class=3D"MsoNormal">Hui Yang (Beijing University Post and Telecommunicat=
ion (BUPT), China)
<o:p></o:p></p>
<p class=3D"MsoNormal">Yang Richard Yang (Yale University, USA)<o:p></o:p><=
/p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Workshop Organizing Committee<o:p></o:p></p>
<p class=3D"MsoNormal">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<o:p></o:p></p>
<p class=3D"MsoNormal">Young Lee, Huawei Technologies (leeyoung (at) huawei=
.com) <o:p>
</o:p></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Luis M. Contreras, Telef=F3nica I&=
#43;D (lmcm (at) tid.es)
<o:p></o:p></span></p>
<p class=3D"MsoNormal">Andrea Fumagalli, The University of Texas at Dallas =
(andreaf (at) utdallas.edu)<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal"><span lang=3D"ES">_______________<o:p></o:p></span><=
/p>
<p class=3D"MsoNormal"><span lang=3D"ES">Luis M. Contreras<o:p></o:p></span=
></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Telef=F3nica I&#43;D<o:p></o:p></s=
pan></p>
<p class=3D"MsoNormal"><span lang=3D"ES"><a href=3D"mailto:lmcm@tid.es">lmc=
m@tid.es</a><o:p></o:p></span></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<br>
<hr>
<font face=3D"Arial" color=3D"Gray" size=3D"1">Este mensaje se dirige exclu=
sivamente a su destinatario. Puede consultar nuestra pol=EDtica de env=EDo =
y recepci=F3n de correo electr=F3nico en el enlace situado m=E1s abajo.<br>
This message is intended exclusively for its addressee. We only send and re=
ceive email on the basis of the terms set out at<br>
http://www.tid.es/ES/PAGINAS/disclaimer.aspx<br>
</font>
</body>
</html>

--Boundary_(ID_Gc6q6bJGX2z8M3SCETU77g)--

From vkg@bell-labs.com  Tue Feb  7 09:10:05 2012
Return-Path: <vkg@bell-labs.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4959921F8876 for <alto@ietfa.amsl.com>; Tue,  7 Feb 2012 09:10:05 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -106.811
X-Spam-Level: 
X-Spam-Status: No, score=-106.811 tagged_above=-999 required=5 tests=[AWL=-0.212, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id kxQjZw4pNFRP for <alto@ietfa.amsl.com>; Tue,  7 Feb 2012 09:10:01 -0800 (PST)
Received: from ihemail4.lucent.com (ihemail4.lucent.com [135.245.0.39]) by ietfa.amsl.com (Postfix) with ESMTP id CCCDD21F8872 for <alto@ietf.org>; Tue,  7 Feb 2012 09:09:58 -0800 (PST)
Received: from usnavsmail2.ndc.alcatel-lucent.com (usnavsmail2.ndc.alcatel-lucent.com [135.3.39.10]) by ihemail4.lucent.com (8.13.8/IER-o) with ESMTP id q17H9sSn021463 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for <alto@ietf.org>; Tue, 7 Feb 2012 11:09:54 -0600 (CST)
Received: from umail.lucent.com (umail-ce2.ndc.lucent.com [135.3.40.63]) by usnavsmail2.ndc.alcatel-lucent.com (8.14.3/8.14.3/GMO) with ESMTP id q17H9s1c016524 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for <alto@ietf.org>; Tue, 7 Feb 2012 11:09:54 -0600
Received: from shoonya.ih.lucent.com (shoonya.ih.lucent.com [135.185.238.235]) by umail.lucent.com (8.13.8/TPES) with ESMTP id q17H9s3D000232 for <alto@ietf.org>; Tue, 7 Feb 2012 11:09:54 -0600 (CST)
Message-ID: <4F315BE2.8070800@bell-labs.com>
Date: Tue, 07 Feb 2012 11:14:10 -0600
From: "Vijay K. Gurbani" <vkg@bell-labs.com>
Organization: Bell Laboratories, Alcatel-Lucent
User-Agent: Mozilla/5.0 (X11; Linux i686; rv:9.0) Gecko/20111222 Thunderbird/9.0
MIME-Version: 1.0
To: alto <alto@ietf.org>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-Scanned-By: MIMEDefang 2.57 on 135.245.2.39
X-Scanned-By: MIMEDefang 2.64 on 135.3.39.10
Subject: [alto] Call for participation: altonext
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 07 Feb 2012 17:10:05 -0000

Folks: This is a reminder on protocol extension discussion
kickoff on the new altoext mailing list (list details in email
below).  There is a community of interest forming on the altoext
mailing list and feedback from people involved in the ALTO
protocol work will be surely useful and certainly welcome.

Thank you.

-------- Original Message --------
Subject: ALTO extensions: new list + problem statement
Date: Mon, 30 Jan 2012 16:54:49 +0100
From: Enrico Marocco <enrico.marocco at telecomitalia.it>
To: alto at ietf.org <alto at ietf.org>

Hi folks,

the new altoext at ietf.org mailing list has just been created to
discuss possible extensions to the about-to-be-finished ALTO protocol.

We have also recently submitted a draft trying to summarize the
previous discussions on the subject:
http://tools.ietf.org/html/draft-marocco-alto-next

In a nutshell: the possible extensions, primarily intended to better
address the CDN/datacenter and server-to-server use cases, mainly
concern:
   + a server-to-client notification mechanism;
   + additional types of information, such as:
     o network link load (average, over some recent timeframe);
     o server load (average, over some recent timeframe);
     o availability of resources such as memory, content, installed
       applications.

The new mailing list has been created in order to get broader feedback
and to get together people with experience in different areas. If you
have an opinion on whether and why the proposed (and possibly further)
extensions will be a good or a bad idea, you can join the discussion
at: https://www.ietf.org/mailman/listinfo/altoext

-- 

- vijay
-- 
Vijay K. Gurbani, Bell Laboratories, Alcatel-Lucent
1960 Lucent Lane, Rm. 9C-533, Naperville, Illinois 60566 (USA)
Email: vkg@{bell-labs.com,acm.org} / vijay.gurbani@alcatel-lucent.com
Web:   http://ect.bell-labs.com/who/vkg/

From lmcm@tid.es  Fri Feb 17 00:20:01 2012
Return-Path: <lmcm@tid.es>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 11F8321E8017; Fri, 17 Feb 2012 00:20:01 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.495
X-Spam-Level: 
X-Spam-Status: No, score=-3.495 tagged_above=-999 required=5 tests=[AWL=0.503,  BAYES_50=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ud3J1Ulpx0mM; Fri, 17 Feb 2012 00:19:59 -0800 (PST)
Received: from correo-bck.tid.es (correo-bck.tid.es [195.235.93.200]) by ietfa.amsl.com (Postfix) with ESMTP id 50CB421F87A3; Fri, 17 Feb 2012 00:19:57 -0800 (PST)
Received: from sbrightmailg02.hi.inet (Sbrightmailg02.hi.inet [10.95.78.105]) by tid.hi.inet (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0LZJ00IJS34UCU@tid.hi.inet>; Fri, 17 Feb 2012 09:19:42 +0100 (MET)
Received: from vanvan (vanvan.hi.inet [10.95.78.49])	by sbrightmailg02.hi.inet (Symantec Messaging Gateway) with SMTP id 7B.B6.02643.E9D0E3F4; Fri, 17 Feb 2012 09:19:42 +0100 (CET)
Received: from correo.tid.es (mailhost.hi.inet [10.95.64.100]) by tid.hi.inet (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTPS id <0LZJ00IJN34UCU@tid.hi.inet>; Fri, 17 Feb 2012 09:19:42 +0100 (MET)
Received: from EXCLU2K7.hi.inet ([10.95.67.65]) by htcasmad1.hi.inet ([192.168.0.1]) with mapi; Fri, 17 Feb 2012 09:19:41 +0100
Date: Fri, 17 Feb 2012 09:19:41 +0100
From: LUIS MIGUEL CONTRERAS MURILLO <lmcm@tid.es>
To: "dc@ietf.org" <dc@ietf.org>, "sdnp@lucidvision.com" <sdnp@lucidvision.com>, "cross-stratum-optimization@googlegroups.com" <cross-stratum-optimization@googlegroups.com>, "alto@ietf.org" <alto@ietf.org>
Message-id: <B348B152E5F11640B2247E54304E53FC590D6FBD2B@EXCLU2K7.hi.inet>
MIME-version: 1.0
Content-type: multipart/alternative; boundary="Boundary_(ID_d/vUFPTbXGJpUf8QYj/C2w)"
Content-language: es-ES
Accept-Language: es-ES, en-US
Thread-topic: CfP "International Workshop on Cross-Stratum Optimization for Cloud Computing and Distributed Networked Applications"
Thread-index: AcztTLBVtMgKlX5cThij1k/V291UbQ==
acceptlanguage: es-ES, en-US
X-AuditID: 0a5f4e69-b7f6b6d000000a53-09-4f3e0d9ee884
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFprHKsWRmVeSWpSXmKPExsXCFe9nqDuP187foGutpMXD7XtZLVrO32V1 YPJYsuQnUwBjFJdNSmpOZllqkb5dAldGy4se1oJzExkrFn/ayNrA+Kyui5GTQ0LAROLGrfuM ELaYxIV769m6GLk4hAS2MUr0bVzKCOE8ZZSYM3cuWJWQQCOjxMofPiA2i4CqxMKr29lAbDYB Q4lZOyexgjQICzQxSuw7d4AJxBERuMAosWbTWVaQKl4BT4mFH94wQdiCEj8m32MBsZkFciW2 tP5jhrDFJeb8mghWzyggK7Hy/GlGiEHtjBKHrrwHaxYR0JOYNH0jVJGMxP/le1kgnhCQWLLn PDOELSrx8vE/1gmMwrOQ7JuFZN8sJPsgbD2JG1OnsEHY2hLLFr6GqtGVmPHvEAuy+AJG9lWM YsVJRZnpGSW5iZk56QZGehmZepl5qSWbGCERlLmDcflOlUOMAhyMSjy8Dgm2/kKsiWXFlbmH GCU5mJREeWfy2PkL8SXlp1RmJBZnxBeV5qQWH2KU4GBWEuF9/hionDclsbIqtSgfJiXDwaEk wbsYpE2wKDU9tSItMweYJmDSTBycIO08QO3dIDW8xQWJucWZ6RD5U4yqHC3Pn19gFGLJy89L lRLnbQQpEgApyijNg5vzilEc6GBh3jkgWR5gooOb8ApoOBPQcPMXViDDSxIRUlINjEtm32AX XxjQfkJWiUfJ+9+OiqXN0v8jfxzYJ3/TrqNFR4Hr0mWRVyk6h9+ee78r/OZLYWOXnu2nF+5w nPGr8Kzo1KR1zjNbltyd1pge5HTT4saWf+qL90wTv6S6f3FU2JOo+8fu9C88cktxt+XZhJSk c2/Lj/dtei1XZTEjxv3JZW3hKUY9iyWUWIozEg21mIuKEwFlMuq7MQMAAA==
Subject: [alto] CfP "International Workshop on Cross-Stratum Optimization for Cloud Computing and Distributed Networked Applications"
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 17 Feb 2012 08:20:01 -0000

--Boundary_(ID_d/vUFPTbXGJpUf8QYj/C2w)
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: quoted-printable

[Apologies if you receive multiple copies of this message]

After receiving several request the organizers have decided to extend the d=
eadline.
Authors are encouraged to register their paper today by indicating the titl=
e and a tentative abstract.
The full paper can be submitted at a later time, but on or before March 2nd

Deadline Extension
Call for Papers for the "International Workshop on Cross-Stratum Optimizati=
on for Cloud Computing and Distributed Networked Applications"
http://cccso.net/ or http://www.cccso.net/

Co-located with the 10th IEEE International Symposium on Parallel and Distr=
ibuted Processing with Applications, ISPA 2012 http://www.arcos.inf.uc3m.es=
/ispa12/index.shtml
July 10-13, 2012

Leganes, Madrid, Spain

Aims and Scope
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
The current lack of interaction between networked applications and the unde=
rlying network during service provisioning can cause inefficiencies in the =
use of network resources and can negatively impact the quality received by =
the final consumers of those applications.
Typical networked applications are offered through Information Technology (=
IT) resources (such as computing and storage facilities) residing in data c=
enters. Data centers then provide the physical and virtual infrastructure i=
n which applications and services are provided. Since the data centers are =
usually distributed geographically around a network, many decisions made in=
 the control and management of application services, such as where to insta=
ntiate another service instance, or which data center out of several is ass=
igned to a new customer, can have a significant impact on the state of the =
network. In the same way, the capabilities and state of the network can hav=
e a major impact on application performance.
Cross-stratum optimization (CSO) is defined as the combined optimization of=
 both the IT data center and the network components of an application that =
aims to provide joint resource optimization, responsiveness to quickly chan=
ging demands from/to application to/from network, enhanced service resilien=
ce using cooperative recovery techniques, and quality of experience assuran=
ce by a better use of existing network and application resources, among oth=
ers.
The CSO involves the overall optimization of application layer (IT) and net=
work resources by envisioning next generation architecture for interactions=
 and exchanges between the two layers to improve service continuity, perfor=
mance guarantees, scalability and manageability. The goal of this workshop =
is to promote the research interest on the optimal integration of applicati=
on and network resources.
This workshop aims to explore the challenges and issues faced by cloud comp=
uting and data center integration with networks. Among the key areas of inv=
estigation to be discussed in the workshop are as follows:
.- Application/network integration architectures and subsystems
.- Use cases, business models and requirements for application/network inte=
gration
.- Control/management issues for application/network integration
.- Network virtualization and its impact for application/network integratio=
n
.- Network-aware application/cloud computing
.- Flexible and scalable networking solutions for distributed Data Centers
.- Joint application/network reliability and security
.- Experimental/trial experience
.- Scalability
.- Joint/shared performance and fault monitoring
.- Multi-domain issues

Important Dates
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
Extended Submission Deadline: 02 March 2012 (*extended final* deadline)
Paper Acceptance Notification: 30 March 2012
Camera-ready Paper Submissions: 13 April 2012
Tentative workshop day: 10 July 2012 (to be confirmed)

Venue
=3D=3D=3D=3D=3D=3D
Universidad Carlos III de Madrid, Leganes, Madrid, Spain

Submision guidelines
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
Target papers should describe original and unpublished work. All papers mus=
t be submitted in PDF format. Authors should submit full papers of up to 6 =
pages, following strictly the IEEE Computer Society Proceedings Manuscript =
style (available at http://www.computer.org/portal/web/cscps/formatting), u=
sing two-column, single-space format, with 10-point font size. Figures and =
references must be included in the 6 pages.
The submission process will be done online by using EasyChair submission sy=
stem:
http://www.easychair.org/conferences/?conf=3Dispa2012
choosing this workshop among the various workshops held in conjunction with=
 ISPA-2012
Submissions received after deadline, exceeding length limit, or not followi=
ng the specified format will not be considered.
The proceedings will be published by the IEEE in the same volume as the mai=
n conference and will be made online through the IEEE Xplore.

Technical Program Committee
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D
Richard Alimi (Google, USA)
Greg Bernstein (Grotto Networking, USA)
TaeSang Choi (ETRI, Korea)
Nicola Ciulli (Nextworks, Italy)
Oscar Gonzalez de Dios (Telef=F3nica I+D, Spain)
Volker Hilt (Bell Labs, USA)
Giada Landi (Nextworks, Italy)
Dan Li (Huawei, China)
Vishwas Manral (HP, USA)
Thomas D. Nadeau (CA Technologies, USA)
Kohei Shiomoto (NTT, Japan)
Ning So (Verizon, USA)
Hui Yang (Beijing University Post and Telecommunication (BUPT), China)
Yang Richard Yang (Yale University, USA)

Workshop Organizing Committee
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
Young Lee, Huawei Technologies (leeyoung (at) huawei.com)
Luis M. Contreras, Telef=F3nica I+D (lmcm (at) tid.es)
Andrea Fumagalli, The University of Texas at Dallas (andreaf (at) utdallas.=
edu)

Best regards,
Luis

_______________
Luis M. Contreras
Telef=F3nica I+D
lmcm@tid.es


________________________________
Este mensaje se dirige exclusivamente a su destinatario. Puede consultar nu=
estra pol=EDtica de env=EDo y recepci=F3n de correo electr=F3nico en el enl=
ace situado m=E1s abajo.
This message is intended exclusively for its addressee. We only send and re=
ceive email on the basis of the terms set out at
http://www.tid.es/ES/PAGINAS/disclaimer.aspx

--Boundary_(ID_d/vUFPTbXGJpUf8QYj/C2w)
Content-type: text/html; charset=iso-8859-1
Content-transfer-encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:=
//www.w3.org/TR/REC-html40">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Diso-8859-=
1">
<meta name=3D"Generator" content=3D"Microsoft Word 14 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0cm;
	margin-bottom:.0001pt;
	font-size:11.0pt;
	font-family:"Calibri","sans-serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EstiloCorreo17
	{mso-style-type:personal-compose;
	font-family:"Calibri","sans-serif";
	color:windowtext;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri","sans-serif";}
@page WordSection1
	{size:612.0pt 792.0pt;
	margin:70.85pt 3.0cm 70.85pt 3.0cm;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal">[Apologies if you receive multiple copies of this me=
ssage]<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">After receiving several request the organizers have =
decided to extend the deadline.<o:p></o:p></p>
<p class=3D"MsoNormal">Authors are encouraged to register their paper today=
 by indicating the title and a tentative abstract.<o:p></o:p></p>
<p class=3D"MsoNormal">The full paper can be submitted at a later time, but=
 on or before March 2nd<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Deadline Extension <o:p></o:p></p>
<p class=3D"MsoNormal">Call for Papers for the &quot;International Workshop=
 on Cross-Stratum Optimization for Cloud Computing and Distributed Networke=
d Applications&quot;<o:p></o:p></p>
<p class=3D"MsoNormal">http://cccso.net/ or http://www.cccso.net/<o:p></o:p=
></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Co-located with the 10th IEEE International Symposiu=
m on Parallel and Distributed Processing with Applications, ISPA 2012 http:=
//www.arcos.inf.uc3m.es/ispa12/index.shtml<o:p></o:p></p>
<p class=3D"MsoNormal">July 10-13, 2012<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Leganes, Madrid, Spain<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Aims and Scope<o:p></o:p></p>
<p class=3D"MsoNormal">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<o:p></o:p></=
p>
<p class=3D"MsoNormal">The current lack of interaction between networked ap=
plications and the underlying network during service provisioning can cause=
 inefficiencies in the use of network resources and can negatively impact t=
he quality received by the final consumers
 of those applications.<o:p></o:p></p>
<p class=3D"MsoNormal">Typical networked applications are offered through I=
nformation Technology (IT) resources (such as computing and storage facilit=
ies) residing in data centers. Data centers then provide the physical and v=
irtual infrastructure in which applications
 and services are provided. Since the data centers are usually distributed =
geographically around a network, many decisions made in the control and man=
agement of application services, such as where to instantiate another servi=
ce instance, or which data center
 out of several is assigned to a new customer, can have a significant impac=
t on the state of the network. In the same way, the capabilities and state =
of the network can have a major impact on application performance.<o:p></o:=
p></p>
<p class=3D"MsoNormal">Cross-stratum optimization (CSO) is defined as the c=
ombined optimization of both the IT data center and the network components =
of an application that aims to provide joint resource optimization, respons=
iveness to quickly changing demands
 from/to application to/from network, enhanced service resilience using coo=
perative recovery techniques, and quality of experience assurance by a bett=
er use of existing network and application resources, among others.<o:p></o=
:p></p>
<p class=3D"MsoNormal">The CSO involves the overall optimization of applica=
tion layer (IT) and network resources by envisioning next generation archit=
ecture for interactions and exchanges between the two layers to improve ser=
vice continuity, performance guarantees,
 scalability and manageability. The goal of this workshop is to promote the=
 research interest on the optimal integration of application and network re=
sources.<o:p></o:p></p>
<p class=3D"MsoNormal">This workshop aims to explore the challenges and iss=
ues faced by cloud computing and data center integration with networks. Amo=
ng the key areas of investigation to be discussed in the workshop are as fo=
llows:<o:p></o:p></p>
<p class=3D"MsoNormal">.- Application/network integration architectures and=
 subsystems<o:p></o:p></p>
<p class=3D"MsoNormal">.- Use cases, business models and requirements for a=
pplication/network integration<o:p></o:p></p>
<p class=3D"MsoNormal">.- Control/management issues for application/network=
 integration<o:p></o:p></p>
<p class=3D"MsoNormal">.- Network virtualization and its impact for applica=
tion/network integration<o:p></o:p></p>
<p class=3D"MsoNormal">.- Network-aware application/cloud computing<o:p></o=
:p></p>
<p class=3D"MsoNormal">.- Flexible and scalable networking solutions for di=
stributed Data Centers<o:p></o:p></p>
<p class=3D"MsoNormal">.- Joint application/network reliability and securit=
y<o:p></o:p></p>
<p class=3D"MsoNormal">.- Experimental/trial experience<o:p></o:p></p>
<p class=3D"MsoNormal">.- Scalability<o:p></o:p></p>
<p class=3D"MsoNormal">.- Joint/shared performance and fault monitoring<o:p=
></o:p></p>
<p class=3D"MsoNormal">.- Multi-domain issues<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Important Dates<o:p></o:p></p>
<p class=3D"MsoNormal">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<o:p></=
o:p></p>
<p class=3D"MsoNormal">Extended Submission Deadline: 02 March 2012 (*extend=
ed final* deadline)<o:p></o:p></p>
<p class=3D"MsoNormal">Paper Acceptance Notification: 30 March 2012 <o:p></=
o:p></p>
<p class=3D"MsoNormal">Camera-ready Paper Submissions: 13 April 2012 <o:p><=
/o:p></p>
<p class=3D"MsoNormal">Tentative workshop day: 10 July 2012 (to be confirme=
d)<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Venue<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"ES">=3D=3D=3D=3D=3D=3D<o:p></o:p></spa=
n></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Universidad Carlos III de Madrid, =
Leganes, Madrid, Spain<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"ES"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal">Submision guidelines<o:p></o:p></p>
<p class=3D"MsoNormal">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D<o:p></o:p></p>
<p class=3D"MsoNormal">Target papers should describe original and unpublish=
ed work. All papers must be submitted in PDF format. Authors should submit =
full papers of up to 6 pages, following strictly the IEEE Computer Society =
Proceedings Manuscript style (available
 at http://www.computer.org/portal/web/cscps/formatting), using two-column,=
 single-space format, with 10-point font size. Figures and references must =
be included in the 6 pages.
<o:p></o:p></p>
<p class=3D"MsoNormal">The submission process will be done online by using =
EasyChair submission system:<o:p></o:p></p>
<p class=3D"MsoNormal">http://www.easychair.org/conferences/?conf=3Dispa201=
2<o:p></o:p></p>
<p class=3D"MsoNormal">choosing this workshop among the various workshops h=
eld in conjunction with ISPA-2012
<o:p></o:p></p>
<p class=3D"MsoNormal">Submissions received after deadline, exceeding lengt=
h limit, or not following the specified format will not be considered.<o:p>=
</o:p></p>
<p class=3D"MsoNormal">The proceedings will be published by the IEEE in the=
 same volume as the main conference and will be made online through the IEE=
E Xplore.<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Technical Program Committee<o:p></o:p></p>
<p class=3D"MsoNormal">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<o:p></o:p></p>
<p class=3D"MsoNormal">Richard Alimi (Google, USA)<o:p></o:p></p>
<p class=3D"MsoNormal">Greg Bernstein (Grotto Networking, USA)<o:p></o:p></=
p>
<p class=3D"MsoNormal"><span lang=3D"ES">TaeSang Choi (ETRI, Korea)<o:p></o=
:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Nicola Ciulli (Nextworks, Italy)<o=
:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Oscar Gonzalez de Dios (Telef=F3ni=
ca I&#43;D, Spain)
<o:p></o:p></span></p>
<p class=3D"MsoNormal">Volker Hilt (Bell Labs, USA) <o:p></o:p></p>
<p class=3D"MsoNormal">Giada Landi (Nextworks, Italy) <o:p></o:p></p>
<p class=3D"MsoNormal">Dan Li (Huawei, China) <o:p></o:p></p>
<p class=3D"MsoNormal">Vishwas Manral (HP, USA)<o:p></o:p></p>
<p class=3D"MsoNormal">Thomas D. Nadeau (CA Technologies, USA) <o:p></o:p><=
/p>
<p class=3D"MsoNormal">Kohei Shiomoto (NTT, Japan) <o:p></o:p></p>
<p class=3D"MsoNormal">Ning So (Verizon, USA) <o:p></o:p></p>
<p class=3D"MsoNormal">Hui Yang (Beijing University Post and Telecommunicat=
ion (BUPT), China)
<o:p></o:p></p>
<p class=3D"MsoNormal">Yang Richard Yang (Yale University, USA)<o:p></o:p><=
/p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Workshop Organizing Committee<o:p></o:p></p>
<p class=3D"MsoNormal">=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<o:p></o:p></p>
<p class=3D"MsoNormal">Young Lee, Huawei Technologies (leeyoung (at) huawei=
.com) <o:p>
</o:p></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Luis M. Contreras, Telef=F3nica I&=
#43;D (lmcm (at) tid.es)
<o:p></o:p></span></p>
<p class=3D"MsoNormal">Andrea Fumagalli, The University of Texas at Dallas =
(andreaf (at) utdallas.edu)<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Best regards,<o:p></o:p></span></p=
>
<p class=3D"MsoNormal"><span lang=3D"ES">Luis<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"ES"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"ES">_______________<o:p></o:p></span><=
/p>
<p class=3D"MsoNormal"><span lang=3D"ES">Luis M. Contreras<o:p></o:p></span=
></p>
<p class=3D"MsoNormal"><span lang=3D"ES">Telef=F3nica I&#43;D<o:p></o:p></s=
pan></p>
<p class=3D"MsoNormal"><span lang=3D"ES">lmcm@tid.es<o:p></o:p></span></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<br>
<hr>
<font face=3D"Arial" color=3D"Gray" size=3D"1">Este mensaje se dirige exclu=
sivamente a su destinatario. Puede consultar nuestra pol=EDtica de env=EDo =
y recepci=F3n de correo electr=F3nico en el enlace situado m=E1s abajo.<br>
This message is intended exclusively for its addressee. We only send and re=
ceive email on the basis of the terms set out at<br>
http://www.tid.es/ES/PAGINAS/disclaimer.aspx<br>
</font>
</body>
</html>

--Boundary_(ID_d/vUFPTbXGJpUf8QYj/C2w)--

From sreekanth.madhavan@huawei.com  Tue Feb 21 21:24:15 2012
Return-Path: <sreekanth.madhavan@huawei.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1401121E8067 for <alto@ietfa.amsl.com>; Tue, 21 Feb 2012 21:24:15 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.599
X-Spam-Level: 
X-Spam-Status: No, score=-6.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id oKl0sXDtmwuK for <alto@ietfa.amsl.com>; Tue, 21 Feb 2012 21:24:14 -0800 (PST)
Received: from szxga04-in.huawei.com (szxga04-in.huawei.com [119.145.14.67]) by ietfa.amsl.com (Postfix) with ESMTP id 67B9221E8065 for <alto@ietf.org>; Tue, 21 Feb 2012 21:24:13 -0800 (PST)
Received: from huawei.com (szxga04-in [172.24.2.12]) by szxga04-in.huawei.com (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0LZS005S94CCKT@szxga04-in.huawei.com> for alto@ietf.org; Wed, 22 Feb 2012 13:24:12 +0800 (CST)
Received: from szxrg02-dlp.huawei.com ([172.24.2.119]) by szxga04-in.huawei.com (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0LZS00KFM4CCII@szxga04-in.huawei.com> for alto@ietf.org; Wed, 22 Feb 2012 13:24:12 +0800 (CST)
Received: from szxeml210-edg.china.huawei.com ([172.24.2.119]) by szxrg02-dlp.huawei.com (MOS 4.1.9-GA)	with ESMTP id AHH45736; Wed, 22 Feb 2012 13:23:59 +0800
Received: from SZXEML419-HUB.china.huawei.com (10.82.67.158) by szxeml210-edg.china.huawei.com (172.24.2.183) with Microsoft SMTP Server (TLS) id 14.1.323.3; Wed, 22 Feb 2012 13:23:45 +0800
Received: from blrprnc03ns (10.18.96.92) by szxeml419-hub.china.huawei.com (10.82.67.158) with Microsoft SMTP Server id 14.1.323.3; Wed, 22 Feb 2012 13:24:37 +0800
Date: Wed, 22 Feb 2012 10:53:49 +0530
From: sreekanth madhavan <sreekanth.madhavan@huawei.com>
X-Originating-IP: [10.18.96.92]
To: alto@ietf.org
Message-id: <000c01ccf122$29365370$7ba2fa50$@com>
MIME-version: 1.0
X-Mailer: Microsoft Office Outlook 12.0
Content-type: text/plain; charset=utf-8
Content-language: en-us
Content-transfer-encoding: 7BIT
Thread-index: AczxHrQcrqRYhLZWSNy1m13yJqiwfgAAni0Q
X-CFilter-Loop: Reflected
Subject: [alto] FW: New Version Notification for draft-alto-caching-subscription-00.txt
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 Feb 2012 05:27:16 -0000

Hi All,

We have submitted a new draft related to Alto caching and subscription. We felt that the existing HTTP based caching mechanisms are not suitable for caching Alto map information.

We also proposed a new mechanism for subscription and notification which we think will help reduce redundant traffic between alto nodes.

Your comments are highly welcome.

http://www.ietf.org/id/draft-alto-caching-subscription-00.txt


Regards,
Sreekanth Madhavan, on behalf of all authors

-----Original Message-----
From: internet-drafts@ietf.org [mailto:internet-drafts@ietf.org] 
Sent: Wednesday, February 22, 2012 10:29 AM
To: sreekanth.madhavan@huawei.com
Cc: sreekanth.madhavan@huawei.com; ravi.nandiraju@huawei.com
Subject: New Version Notification for draft-alto-caching-subscription-00.txt

A new version of I-D, draft-alto-caching-subscription-00.txt has been successfully submitted by Sreekanth Madhavan and posted to the IETF repository.

Filename:	 draft-alto-caching-subscription
Revision:	 00
Title:		 ALTO Caching and Subscription
Creation date:	 2012-02-22
WG ID:		 Individual Submission
Number of pages: 19

Abstract:
   The specification of the ALTO protocol uses map based approaches
   assuming that the information provided is static for a longer period
   of time.  But in some cases network operators reallocate IP subnets
   from time to time which in turn changes the mapping partitions[I-
   D.ietf-alto-deployments].  Since the ALTO clients are unaware of the
   map information changes, clients need to query the servers for every
   service request and many such requests are redundant because the
   information was not changed.  The purpose of this memo is to provide
   two mechanisms which will help the ALTO clients to be informed of the
   map information changes. a) ALTO clients cache the map information
   and the servers provide the expiration time for invalidation. b) ALTO
   clients can subscribe for event notifications from the server

                                                                                  


The IETF Secretariat


From richard.alimi@gmail.com  Tue Feb 21 22:44:53 2012
Return-Path: <richard.alimi@gmail.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CE8B721F8747 for <alto@ietfa.amsl.com>; Tue, 21 Feb 2012 22:44:53 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.827
X-Spam-Level: 
X-Spam-Status: No, score=-2.827 tagged_above=-999 required=5 tests=[AWL=0.150,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id I-fjTMS+CpLN for <alto@ietfa.amsl.com>; Tue, 21 Feb 2012 22:44:41 -0800 (PST)
Received: from mail-iy0-f172.google.com (mail-iy0-f172.google.com [209.85.210.172]) by ietfa.amsl.com (Postfix) with ESMTP id 7B3FA21F8622 for <alto@ietf.org>; Tue, 21 Feb 2012 22:44:41 -0800 (PST)
Received: by iagf6 with SMTP id f6so12473759iag.31 for <alto@ietf.org>; Tue, 21 Feb 2012 22:44:41 -0800 (PST)
Received-SPF: pass (google.com: domain of richard.alimi@gmail.com designates 10.50.178.73 as permitted sender) client-ip=10.50.178.73; 
Authentication-Results: mr.google.com; spf=pass (google.com: domain of richard.alimi@gmail.com designates 10.50.178.73 as permitted sender) smtp.mail=richard.alimi@gmail.com; dkim=pass header.i=richard.alimi@gmail.com
Received: from mr.google.com ([10.50.178.73]) by 10.50.178.73 with SMTP id cw9mr24760422igc.23.1329893081197 (num_hops = 1); Tue, 21 Feb 2012 22:44:41 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:cc:content-type :content-transfer-encoding; bh=YsI1iewWMqbycnLCfhcMHRj+aBSeirrNBGU5tfiQOgs=; b=G+jqvwBd6KMh0WouwRfdf/7+mUnx9+dlqGvfPEOHrTMCnF4YKUCR183/QPmALkP0ki NcLQRHSekRwL8ZqmAkU1Ga3nAK/E+sQbKSnjZmw3yS4BkkkvAEhwjuvuDhBRILiYcgQ1 CYEFfR7vJl/bfs0kOTTG1gvg1l94ggCaJbnQM=
Received: by 10.50.178.73 with SMTP id cw9mr20057474igc.23.1329893081140; Tue, 21 Feb 2012 22:44:41 -0800 (PST)
MIME-Version: 1.0
Sender: richard.alimi@gmail.com
Received: by 10.231.184.84 with HTTP; Tue, 21 Feb 2012 22:44:21 -0800 (PST)
In-Reply-To: <000c01ccf122$29365370$7ba2fa50$@com>
References: <000c01ccf122$29365370$7ba2fa50$@com>
From: Richard Alimi <rich@velvetsea.net>
Date: Tue, 21 Feb 2012 22:44:21 -0800
X-Google-Sender-Auth: aYj7TJWmZ_PQF1Iwhdh_S36HJ0Y
Message-ID: <CA+cvDaY0eqQ1S7RXXrfqPmX8SyaLVazL3hPcXQbSzBkXy639Vg@mail.gmail.com>
To: sreekanth madhavan <sreekanth.madhavan@huawei.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: alto@ietf.org
Subject: Re: [alto] FW: New Version Notification for draft-alto-caching-subscription-00.txt
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 Feb 2012 06:44:53 -0000

Thank you for submitting the draft.  Reading through very quickly,
there are a couple of comments:
(1) Note that HTTP already allows expiration date/time to be conveyed
for the entire document. I don't see a reason for adding another way
to convey that. (Yes, the new thing discussed in your draft is caching
a subset of the content.)
(2) There are ways to do caching in HTTP for POST requests (e.g.,
server can reply with an alternate location and the client does a GET
on that location).
(3) Reading through quickly, anyways, the notification scheme seems
conceptually similar to that discussed in
http://tools.ietf.org/html/draft-sun-alto-notification-02.  There was
extensive discussion on the list for that document, so it may be worth
reading through that (if you haven't already, of course).

Thanks again for submitting the draft,
Rich

On Tue, Feb 21, 2012 at 9:23 PM, sreekanth madhavan
<sreekanth.madhavan@huawei.com> wrote:
> Hi All,
>
> We have submitted a new draft related to Alto caching and subscription. W=
e felt that the existing HTTP based caching mechanisms are not suitable for=
 caching Alto map information.
>
> We also proposed a new mechanism for subscription and notification which =
we think will help reduce redundant traffic between alto nodes.
>
> Your comments are highly welcome.
>
> http://www.ietf.org/id/draft-alto-caching-subscription-00.txt
>
>
> Regards,
> Sreekanth Madhavan, on behalf of all authors
>
> -----Original Message-----
> From: internet-drafts@ietf.org [mailto:internet-drafts@ietf.org]
> Sent: Wednesday, February 22, 2012 10:29 AM
> To: sreekanth.madhavan@huawei.com
> Cc: sreekanth.madhavan@huawei.com; ravi.nandiraju@huawei.com
> Subject: New Version Notification for draft-alto-caching-subscription-00.=
txt
>
> A new version of I-D, draft-alto-caching-subscription-00.txt has been suc=
cessfully submitted by Sreekanth Madhavan and posted to the IETF repository=
.
>
> Filename: =A0 =A0 =A0 =A0draft-alto-caching-subscription
> Revision: =A0 =A0 =A0 =A000
> Title: =A0 =A0 =A0 =A0 =A0 ALTO Caching and Subscription
> Creation date: =A0 2012-02-22
> WG ID: =A0 =A0 =A0 =A0 =A0 Individual Submission
> Number of pages: 19
>
> Abstract:
> =A0 The specification of the ALTO protocol uses map based approaches
> =A0 assuming that the information provided is static for a longer period
> =A0 of time. =A0But in some cases network operators reallocate IP subnets
> =A0 from time to time which in turn changes the mapping partitions[I-
> =A0 D.ietf-alto-deployments]. =A0Since the ALTO clients are unaware of th=
e
> =A0 map information changes, clients need to query the servers for every
> =A0 service request and many such requests are redundant because the
> =A0 information was not changed. =A0The purpose of this memo is to provid=
e
> =A0 two mechanisms which will help the ALTO clients to be informed of the
> =A0 map information changes. a) ALTO clients cache the map information
> =A0 and the servers provide the expiration time for invalidation. b) ALTO
> =A0 clients can subscribe for event notifications from the server
>
>
>
>
> The IETF Secretariat
>
> _______________________________________________
> alto mailing list
> alto@ietf.org
> https://www.ietf.org/mailman/listinfo/alto

From sreekanth.madhavan@huawei.com  Wed Feb 22 06:49:54 2012
Return-Path: <sreekanth.madhavan@huawei.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E712B21F865A for <alto@ietfa.amsl.com>; Wed, 22 Feb 2012 06:49:54 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.246
X-Spam-Level: 
X-Spam-Status: No, score=-6.246 tagged_above=-999 required=5 tests=[AWL=0.353,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wpke8q10THbn for <alto@ietfa.amsl.com>; Wed, 22 Feb 2012 06:49:54 -0800 (PST)
Received: from szxga04-in.huawei.com (szxga04-in.huawei.com [119.145.14.67]) by ietfa.amsl.com (Postfix) with ESMTP id A594621F8709 for <alto@ietf.org>; Wed, 22 Feb 2012 06:49:53 -0800 (PST)
Received: from huawei.com (szxga04-in [172.24.2.12]) by szxga04-in.huawei.com (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0LZS004NWUFU4G@szxga04-in.huawei.com> for alto@ietf.org; Wed, 22 Feb 2012 22:47:54 +0800 (CST)
Received: from szxrg02-dlp.huawei.com ([172.24.2.119]) by szxga04-in.huawei.com (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0LZS00JHKUFUYY@szxga04-in.huawei.com> for alto@ietf.org; Wed, 22 Feb 2012 22:47:54 +0800 (CST)
Received: from szxeml212-edg.china.huawei.com ([172.24.2.119]) by szxrg02-dlp.huawei.com (MOS 4.1.9-GA)	with ESMTP id AHI06566; Wed, 22 Feb 2012 22:47:49 +0800
Received: from SZXEML422-HUB.china.huawei.com (10.82.67.161) by szxeml212-edg.china.huawei.com (172.24.2.181) with Microsoft SMTP Server (TLS) id 14.1.323.3; Wed, 22 Feb 2012 22:47:47 +0800
Received: from blrprnc03ns (10.18.96.92) by szxeml422-hub.china.huawei.com (10.82.67.161) with Microsoft SMTP Server id 14.1.323.3; Wed, 22 Feb 2012 22:47:42 +0800
Date: Wed, 22 Feb 2012 20:17:41 +0530
From: sreekanth madhavan <sreekanth.madhavan@huawei.com>
In-reply-to: <CA+cvDaY0eqQ1S7RXXrfqPmX8SyaLVazL3hPcXQbSzBkXy639Vg@mail.gmail.com>
X-Originating-IP: [10.18.96.92]
To: 'Richard Alimi' <rich@velvetsea.net>, 'sreekanth madhavan' <sreekanth.madhavan@huawei.com>
Message-id: <000301ccf170$eea45500$cbecff00$@com>
MIME-version: 1.0
X-Mailer: Microsoft Office Outlook 12.0
Content-type: text/plain; charset=iso-8859-1
Content-language: en-us
Content-transfer-encoding: quoted-printable
Thread-index: AczxLYoeMLWii8Q+S+SZ/qRVrZ6bSQAQK78g
X-CFilter-Loop: Reflected
References: <000c01ccf122$29365370$7ba2fa50$@com> <CA+cvDaY0eqQ1S7RXXrfqPmX8SyaLVazL3hPcXQbSzBkXy639Vg@mail.gmail.com>
Cc: alto@ietf.org
Subject: Re: [alto] FW: New Version Notification for draft-alto-caching-subscription-00.txt
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 Feb 2012 14:49:55 -0000

Thanks for the review and quick response.

Comments inline.

Regards,
Sreekanth

-----Original Message-----
From: richard.alimi@gmail.com [mailto:richard.alimi@gmail.com] On Behalf =
Of
Richard Alimi
Sent: Wednesday, February 22, 2012 12:14 PM
To: sreekanth madhavan
Cc: alto@ietf.org
Subject: Re: [alto] FW: New Version Notification for
draft-alto-caching-subscription-00.txt

Thank you for submitting the draft.  Reading through very quickly,
there are a couple of comments:
(1) Note that HTTP already allows expiration date/time to be conveyed
for the entire document. I don't see a reason for adding another way
to convey that. (Yes, the new thing discussed in your draft is caching
a subset of the content.)
(2) There are ways to do caching in HTTP for POST requests (e.g.,
server can reply with an alternate location and the client does a GET
on that location).


[SK]
HTTP based caching mechanisms cannot be used for subset of the content. =
In
most cases ISPs change part of the network map information related to =
some
of the PIDs.
In the above case, ALTO clients ideally will be interested only in the
changed information and it does not make sense to resend the whole
information to the clients.

AFAIK, In the ALTO context, HTTP is similar to a transport mechanism. =
The
validity of the map information is related to the ALTO layer and =
delegating
this to the HTTP layer is probably not a good idea.

Regarding the redirect mechanism for POST which you have suggested, it
involves additional messaging which can be avoided by the mechanism =
proposed
in the draft.

(3) Reading through quickly, anyways, the notification scheme seems
conceptually similar to that discussed in
http://tools.ietf.org/html/draft-sun-alto-notification-02.  There was
extensive discussion on the list for that document, so it may be worth
reading through that (if you haven't already, of course).

[SK]
Thanks for providing reference to sun-alto-notification draft. In our
proposal the notification message carries the map information (only the
changed) unlike the referred draft where the client needs to query =
again.=20

Notification framework helps in fast propagation of information changes. =
In
the recent developments in the working group, alto-next also proposes =
server
notification mechanism.

Thanks again for submitting the draft,
Rich

On Tue, Feb 21, 2012 at 9:23 PM, sreekanth madhavan
<sreekanth.madhavan@huawei.com> wrote:
> Hi All,
>
> We have submitted a new draft related to Alto caching and =
subscription. We
felt that the existing HTTP based caching mechanisms are not suitable =
for
caching Alto map information.
>
> We also proposed a new mechanism for subscription and notification =
which
we think will help reduce redundant traffic between alto nodes.
>
> Your comments are highly welcome.
>
> http://www.ietf.org/id/draft-alto-caching-subscription-00.txt
>
>
> Regards,
> Sreekanth Madhavan, on behalf of all authors
>
> -----Original Message-----
> From: internet-drafts@ietf.org [mailto:internet-drafts@ietf.org]
> Sent: Wednesday, February 22, 2012 10:29 AM
> To: sreekanth.madhavan@huawei.com
> Cc: sreekanth.madhavan@huawei.com; ravi.nandiraju@huawei.com
> Subject: New Version Notification for
draft-alto-caching-subscription-00.txt
>
> A new version of I-D, draft-alto-caching-subscription-00.txt has been
successfully submitted by Sreekanth Madhavan and posted to the IETF
repository.
>
> Filename: =A0 =A0 =A0 =A0draft-alto-caching-subscription
> Revision: =A0 =A0 =A0 =A000
> Title: =A0 =A0 =A0 =A0 =A0 ALTO Caching and Subscription
> Creation date: =A0 2012-02-22
> WG ID: =A0 =A0 =A0 =A0 =A0 Individual Submission
> Number of pages: 19
>
> Abstract:
> =A0 The specification of the ALTO protocol uses map based approaches
> =A0 assuming that the information provided is static for a longer =
period
> =A0 of time. =A0But in some cases network operators reallocate IP =
subnets
> =A0 from time to time which in turn changes the mapping partitions[I-
> =A0 D.ietf-alto-deployments]. =A0Since the ALTO clients are unaware of =
the
> =A0 map information changes, clients need to query the servers for =
every
> =A0 service request and many such requests are redundant because the
> =A0 information was not changed. =A0The purpose of this memo is to =
provide
> =A0 two mechanisms which will help the ALTO clients to be informed of =
the
> =A0 map information changes. a) ALTO clients cache the map information
> =A0 and the servers provide the expiration time for invalidation. b) =
ALTO
> =A0 clients can subscribe for event notifications from the server
>
>
>
>
> The IETF Secretariat
>
> _______________________________________________
> alto mailing list
> alto@ietf.org
> https://www.ietf.org/mailman/listinfo/alto


From ben@niven-jenkins.co.uk  Wed Feb 22 07:08:31 2012
Return-Path: <ben@niven-jenkins.co.uk>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A894021F8814 for <alto@ietfa.amsl.com>; Wed, 22 Feb 2012 07:08:31 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.151
X-Spam-Level: 
X-Spam-Status: No, score=-103.151 tagged_above=-999 required=5 tests=[AWL=0.448, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WlVTYXsV4Ttc for <alto@ietfa.amsl.com>; Wed, 22 Feb 2012 07:08:27 -0800 (PST)
Received: from mailex.mailcore.me (mailex.mailcore.me [94.136.40.62]) by ietfa.amsl.com (Postfix) with ESMTP id 644E521F8811 for <alto@ietf.org>; Wed, 22 Feb 2012 07:08:27 -0800 (PST)
Received: from host1.cachelogic.com ([212.44.43.80] helo=xxx.dhcp.cachelogic.com) by mail5.atlas.pipex.net with esmtpa (Exim 4.71) (envelope-from <ben@niven-jenkins.co.uk>) id 1S0Dnu-0001z5-3t; Wed, 22 Feb 2012 15:08:26 +0000
Mime-Version: 1.0 (Apple Message framework v1084)
Content-Type: text/plain; charset=us-ascii
From: Ben Niven-Jenkins <ben@niven-jenkins.co.uk>
In-Reply-To: <000301ccf170$eea45500$cbecff00$@com>
Date: Wed, 22 Feb 2012 15:08:24 +0000
Content-Transfer-Encoding: quoted-printable
Message-Id: <B4AFE7CF-0A03-4F22-ADB7-56A2EFC96E21@niven-jenkins.co.uk>
References: <000c01ccf122$29365370$7ba2fa50$@com> <CA+cvDaY0eqQ1S7RXXrfqPmX8SyaLVazL3hPcXQbSzBkXy639Vg@mail.gmail.com> <000301ccf170$eea45500$cbecff00$@com>
To: sreekanth madhavan <sreekanth.madhavan@huawei.com>
X-Mailer: Apple Mail (2.1084)
X-Mailcore-Auth: 9600544
X-Mailcore-Domain: 172912
Cc: alto@ietf.org
Subject: Re: [alto] FW: New Version Notification for draft-alto-caching-subscription-00.txt
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 Feb 2012 15:08:31 -0000

On 22 Feb 2012, at 14:47, sreekanth madhavan wrote:
> From: richard.alimi@gmail.com [mailto:richard.alimi@gmail.com] On =
Behalf Of
> Richard Alimi
> (2) There are ways to do caching in HTTP for POST requests (e.g.,
> server can reply with an alternate location and the client does a GET
> on that location).
>=20
>=20
> [SK]
> HTTP based caching mechanisms cannot be used for subset of the =
content. In
> most cases ISPs change part of the network map information related to =
some
> of the PIDs.
> In the above case, ALTO clients ideally will be interested only in the
> changed information and it does not make sense to resend the whole
> information to the clients.

A notification mechanism doesn't help you with that problem though. What =
you need is a mechanism to provide incremental updates (i.e. only send =
the things that have changed), just being notified there is a change =
doesn't help with caching.

Ben

>=20
> AFAIK, In the ALTO context, HTTP is similar to a transport mechanism. =
The
> validity of the map information is related to the ALTO layer and =
delegating
> this to the HTTP layer is probably not a good idea.
>=20
> Regarding the redirect mechanism for POST which you have suggested, it
> involves additional messaging which can be avoided by the mechanism =
proposed
> in the draft.
>=20
> (3) Reading through quickly, anyways, the notification scheme seems
> conceptually similar to that discussed in
> http://tools.ietf.org/html/draft-sun-alto-notification-02.  There was
> extensive discussion on the list for that document, so it may be worth
> reading through that (if you haven't already, of course).
>=20
> [SK]
> Thanks for providing reference to sun-alto-notification draft. In our
> proposal the notification message carries the map information (only =
the
> changed) unlike the referred draft where the client needs to query =
again.=20
>=20
> Notification framework helps in fast propagation of information =
changes. In
> the recent developments in the working group, alto-next also proposes =
server
> notification mechanism.
>=20
> Thanks again for submitting the draft,
> Rich
>=20
> On Tue, Feb 21, 2012 at 9:23 PM, sreekanth madhavan
> <sreekanth.madhavan@huawei.com> wrote:
>> Hi All,
>>=20
>> We have submitted a new draft related to Alto caching and =
subscription. We
> felt that the existing HTTP based caching mechanisms are not suitable =
for
> caching Alto map information.
>>=20
>> We also proposed a new mechanism for subscription and notification =
which
> we think will help reduce redundant traffic between alto nodes.
>>=20
>> Your comments are highly welcome.
>>=20
>> http://www.ietf.org/id/draft-alto-caching-subscription-00.txt
>>=20
>>=20
>> Regards,
>> Sreekanth Madhavan, on behalf of all authors
>>=20
>> -----Original Message-----
>> From: internet-drafts@ietf.org [mailto:internet-drafts@ietf.org]
>> Sent: Wednesday, February 22, 2012 10:29 AM
>> To: sreekanth.madhavan@huawei.com
>> Cc: sreekanth.madhavan@huawei.com; ravi.nandiraju@huawei.com
>> Subject: New Version Notification for
> draft-alto-caching-subscription-00.txt
>>=20
>> A new version of I-D, draft-alto-caching-subscription-00.txt has been
> successfully submitted by Sreekanth Madhavan and posted to the IETF
> repository.
>>=20
>> Filename:        draft-alto-caching-subscription
>> Revision:        00
>> Title:           ALTO Caching and Subscription
>> Creation date:   2012-02-22
>> WG ID:           Individual Submission
>> Number of pages: 19
>>=20
>> Abstract:
>>   The specification of the ALTO protocol uses map based approaches
>>   assuming that the information provided is static for a longer =
period
>>   of time.  But in some cases network operators reallocate IP subnets
>>   from time to time which in turn changes the mapping partitions[I-
>>   D.ietf-alto-deployments].  Since the ALTO clients are unaware of =
the
>>   map information changes, clients need to query the servers for =
every
>>   service request and many such requests are redundant because the
>>   information was not changed.  The purpose of this memo is to =
provide
>>   two mechanisms which will help the ALTO clients to be informed of =
the
>>   map information changes. a) ALTO clients cache the map information
>>   and the servers provide the expiration time for invalidation. b) =
ALTO
>>   clients can subscribe for event notifications from the server
>>=20
>>=20
>>=20
>>=20
>> The IETF Secretariat
>>=20
>> _______________________________________________
>> alto mailing list
>> alto@ietf.org
>> https://www.ietf.org/mailman/listinfo/alto
>=20
> _______________________________________________
> alto mailing list
> alto@ietf.org
> https://www.ietf.org/mailman/listinfo/alto


From rpenno@juniper.net  Wed Feb 22 07:35:05 2012
Return-Path: <rpenno@juniper.net>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8BB5621F84FC for <alto@ietfa.amsl.com>; Wed, 22 Feb 2012 07:35:05 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.599
X-Spam-Level: 
X-Spam-Status: No, score=-6.599 tagged_above=-999 required=5 tests=[AWL=0.000,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id O1gWZSj7S0D2 for <alto@ietfa.amsl.com>; Wed, 22 Feb 2012 07:35:01 -0800 (PST)
Received: from exprod7og112.obsmtp.com (exprod7og112.obsmtp.com [64.18.2.177]) by ietfa.amsl.com (Postfix) with ESMTP id 0E15521F84F2 for <alto@ietf.org>; Wed, 22 Feb 2012 07:34:57 -0800 (PST)
Received: from P-EMHUB03-HQ.jnpr.net ([66.129.224.36]) (using TLSv1) by exprod7ob112.postini.com ([64.18.6.12]) with SMTP ID DSNKT0ULIUcalgTcigF6UiW6g4XEDI8KUDn7@postini.com; Wed, 22 Feb 2012 07:35:00 PST
Received: from p-emfe02-wf.jnpr.net (172.28.145.25) by P-EMHUB03-HQ.jnpr.net (172.24.192.37) with Microsoft SMTP Server (TLS) id 8.3.213.0; Wed, 22 Feb 2012 07:33:12 -0800
Received: from EMBX01-WF.jnpr.net ([fe80::1914:3299:33d9:e43b]) by p-emfe02-wf.jnpr.net ([fe80::c126:c633:d2dc:8090%11]) with mapi; Wed, 22 Feb 2012 10:33:11 -0500
From: Reinaldo Penno <rpenno@juniper.net>
To: Ben Niven-Jenkins <ben@niven-jenkins.co.uk>, sreekanth madhavan <sreekanth.madhavan@huawei.com>
Date: Wed, 22 Feb 2012 10:33:08 -0500
Thread-Topic: [alto] FW: New Version Notification for draft-alto-caching-subscription-00.txt
Thread-Index: Aczxc9t+DwskrnJ2QEmjhwsGEjSrwAAA2toP
Message-ID: <CB6A4AB4.63F17%rpenno@juniper.net>
In-Reply-To: <B4AFE7CF-0A03-4F22-ADB7-56A2EFC96E21@niven-jenkins.co.uk>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-Entourage/13.11.0.110726
acceptlanguage: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-EXCLAIMER-MD-CONFIG: e4081efb-6d29-443c-8708-750833aec629
Cc: "alto@ietf.org" <alto@ietf.org>
Subject: Re: [alto] FW: New Version Notification for draft-alto-caching-subscription-00.txt
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 Feb 2012 15:35:05 -0000

On 2/22/12 7:08 AM, "Ben Niven-Jenkins" <ben@niven-jenkins.co.uk> wrote:

>=20
> On 22 Feb 2012, at 14:47, sreekanth madhavan wrote:
>> From: richard.alimi@gmail.com [mailto:richard.alimi@gmail.com] On Behalf=
 Of
>> Richard Alimi
>> (2) There are ways to do caching in HTTP for POST requests (e.g.,
>> server can reply with an alternate location and the client does a GET
>> on that location).
>>=20
>>=20
>> [SK]
>> HTTP based caching mechanisms cannot be used for subset of the content. =
In
>> most cases ISPs change part of the network map information related to so=
me
>> of the PIDs.
>> In the above case, ALTO clients ideally will be interested only in the
>> changed information and it does not make sense to resend the whole
>> information to the clients.
>=20
> A notification mechanism doesn't help you with that problem though. What =
you
> need is a mechanism to provide incremental updates (i.e. only send the th=
ings
> that have changed), just being notified there is a change doesn't help wi=
th
> caching.

Which was planned, written but finally removed from base ALTO protocol. But
irrespective, incremental updates is something I've always seen as a
important piece of the puzzle. But this is something Server and clients can
probably do it almost without changes to base spec as long as the diff rule=
s
are well understood.

>=20
> Ben
>=20
>>=20
>> AFAIK, In the ALTO context, HTTP is similar to a transport mechanism. Th=
e
>> validity of the map information is related to the ALTO layer and delegat=
ing
>> this to the HTTP layer is probably not a good idea.
>>=20
>> Regarding the redirect mechanism for POST which you have suggested, it
>> involves additional messaging which can be avoided by the mechanism prop=
osed
>> in the draft.
>>=20
>> (3) Reading through quickly, anyways, the notification scheme seems
>> conceptually similar to that discussed in
>> http://tools.ietf.org/html/draft-sun-alto-notification-02.  There was
>> extensive discussion on the list for that document, so it may be worth
>> reading through that (if you haven't already, of course).
>>=20
>> [SK]
>> Thanks for providing reference to sun-alto-notification draft. In our
>> proposal the notification message carries the map information (only the
>> changed) unlike the referred draft where the client needs to query again=
.
>>=20
>> Notification framework helps in fast propagation of information changes.=
 In
>> the recent developments in the working group, alto-next also proposes se=
rver
>> notification mechanism.
>>=20
>> Thanks again for submitting the draft,
>> Rich
>>=20
>> On Tue, Feb 21, 2012 at 9:23 PM, sreekanth madhavan
>> <sreekanth.madhavan@huawei.com> wrote:
>>> Hi All,
>>>=20
>>> We have submitted a new draft related to Alto caching and subscription.=
 We
>> felt that the existing HTTP based caching mechanisms are not suitable fo=
r
>> caching Alto map information.
>>>=20
>>> We also proposed a new mechanism for subscription and notification whic=
h
>> we think will help reduce redundant traffic between alto nodes.
>>>=20
>>> Your comments are highly welcome.
>>>=20
>>> http://www.ietf.org/id/draft-alto-caching-subscription-00.txt
>>>=20
>>>=20
>>> Regards,
>>> Sreekanth Madhavan, on behalf of all authors
>>>=20
>>> -----Original Message-----
>>> From: internet-drafts@ietf.org [mailto:internet-drafts@ietf.org]
>>> Sent: Wednesday, February 22, 2012 10:29 AM
>>> To: sreekanth.madhavan@huawei.com
>>> Cc: sreekanth.madhavan@huawei.com; ravi.nandiraju@huawei.com
>>> Subject: New Version Notification for
>> draft-alto-caching-subscription-00.txt
>>>=20
>>> A new version of I-D, draft-alto-caching-subscription-00.txt has been
>> successfully submitted by Sreekanth Madhavan and posted to the IETF
>> repository.
>>>=20
>>> Filename:        draft-alto-caching-subscription
>>> Revision:        00
>>> Title:           ALTO Caching and Subscription
>>> Creation date:   2012-02-22
>>> WG ID:           Individual Submission
>>> Number of pages: 19
>>>=20
>>> Abstract:
>>>   The specification of the ALTO protocol uses map based approaches
>>>   assuming that the information provided is static for a longer period
>>>   of time.  But in some cases network operators reallocate IP subnets
>>>   from time to time which in turn changes the mapping partitions[I-
>>>   D.ietf-alto-deployments].  Since the ALTO clients are unaware of the
>>>   map information changes, clients need to query the servers for every
>>>   service request and many such requests are redundant because the
>>>   information was not changed.  The purpose of this memo is to provide
>>>   two mechanisms which will help the ALTO clients to be informed of the
>>>   map information changes. a) ALTO clients cache the map information
>>>   and the servers provide the expiration time for invalidation. b) ALTO
>>>   clients can subscribe for event notifications from the server
>>>=20
>>>=20
>>>=20
>>>=20
>>> The IETF Secretariat
>>>=20
>>> _______________________________________________
>>> alto mailing list
>>> alto@ietf.org
>>> https://www.ietf.org/mailman/listinfo/alto
>>=20
>> _______________________________________________
>> alto mailing list
>> alto@ietf.org
>> https://www.ietf.org/mailman/listinfo/alto
>=20
> _______________________________________________
> alto mailing list
> alto@ietf.org
> https://www.ietf.org/mailman/listinfo/alto


From richard.alimi@gmail.com  Wed Feb 22 07:43:33 2012
Return-Path: <richard.alimi@gmail.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5DBA221F8717 for <alto@ietfa.amsl.com>; Wed, 22 Feb 2012 07:43:33 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.848
X-Spam-Level: 
X-Spam-Status: No, score=-2.848 tagged_above=-999 required=5 tests=[AWL=0.129,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2MgBv7B4kJqO for <alto@ietfa.amsl.com>; Wed, 22 Feb 2012 07:43:29 -0800 (PST)
Received: from mail-gy0-f172.google.com (mail-gy0-f172.google.com [209.85.160.172]) by ietfa.amsl.com (Postfix) with ESMTP id 05A7121F867F for <alto@ietf.org>; Wed, 22 Feb 2012 07:43:28 -0800 (PST)
Received: by ghbg16 with SMTP id g16so113184ghb.31 for <alto@ietf.org>; Wed, 22 Feb 2012 07:43:28 -0800 (PST)
Received-SPF: pass (google.com: domain of richard.alimi@gmail.com designates 10.50.169.9 as permitted sender) client-ip=10.50.169.9; 
Authentication-Results: mr.google.com; spf=pass (google.com: domain of richard.alimi@gmail.com designates 10.50.169.9 as permitted sender) smtp.mail=richard.alimi@gmail.com; dkim=pass header.i=richard.alimi@gmail.com
Received: from mr.google.com ([10.50.169.9]) by 10.50.169.9 with SMTP id aa9mr27327610igc.23.1329925408463 (num_hops = 1); Wed, 22 Feb 2012 07:43:28 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:cc:content-type :content-transfer-encoding; bh=bfC5jHuGvlQJtkdBUIRuAEuVhGxSIqeDBmYP+Tl8g6g=; b=Rq/juN5l9/BDdQuLt/lOZ8X12CKsRvzctTkq6oxFgrz0hT1z8NiZVpvyJtTkZ4u5+N BSpPO6jGUZKvqS+TuIlg5Sm6UvtyqKdr1WNOU+HiuuxtagdVHIhkMsu5Wy9pzS5yysCD ZqiBdGI+ZKExj9RFMNGQmkFKeZhRvwV1dRUrE=
Received: by 10.50.169.9 with SMTP id aa9mr22157802igc.23.1329925408402; Wed, 22 Feb 2012 07:43:28 -0800 (PST)
MIME-Version: 1.0
Sender: richard.alimi@gmail.com
Received: by 10.231.184.84 with HTTP; Wed, 22 Feb 2012 07:43:08 -0800 (PST)
In-Reply-To: <B4AFE7CF-0A03-4F22-ADB7-56A2EFC96E21@niven-jenkins.co.uk>
References: <000c01ccf122$29365370$7ba2fa50$@com> <CA+cvDaY0eqQ1S7RXXrfqPmX8SyaLVazL3hPcXQbSzBkXy639Vg@mail.gmail.com> <000301ccf170$eea45500$cbecff00$@com> <B4AFE7CF-0A03-4F22-ADB7-56A2EFC96E21@niven-jenkins.co.uk>
From: Richard Alimi <rich@velvetsea.net>
Date: Wed, 22 Feb 2012 07:43:08 -0800
X-Google-Sender-Auth: F7y1dR5USNpIN5j8G1NfULYesBc
Message-ID: <CA+cvDaZMv3v4Vp7yLrSHn7kHt+Ag_Q4kr+MkRGv8cHMqfWYT+Q@mail.gmail.com>
To: Ben Niven-Jenkins <ben@niven-jenkins.co.uk>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: sreekanth madhavan <sreekanth.madhavan@huawei.com>, alto@ietf.org
Subject: Re: [alto] FW: New Version Notification for draft-alto-caching-subscription-00.txt
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 Feb 2012 15:43:33 -0000

Weird - I didn't get Sreekanth's original reply, so replying to Ben's inste=
ad :)

On Wed, Feb 22, 2012 at 7:08 AM, Ben Niven-Jenkins
<ben@niven-jenkins.co.uk> wrote:
>
> On 22 Feb 2012, at 14:47, sreekanth madhavan wrote:
>> From: richard.alimi@gmail.com [mailto:richard.alimi@gmail.com] On Behalf=
 Of
>> Richard Alimi
>> (2) There are ways to do caching in HTTP for POST requests (e.g.,
>> server can reply with an alternate location and the client does a GET
>> on that location).
>>
>>
>> [SK]
>> HTTP based caching mechanisms cannot be used for subset of the content. =
In
>> most cases ISPs change part of the network map information related to so=
me
>> of the PIDs.
>> In the above case, ALTO clients ideally will be interested only in the
>> changed information and it does not make sense to resend the whole
>> information to the clients.
>
> A notification mechanism doesn't help you with that problem though. What =
you need is a mechanism to provide incremental updates (i.e. only send the =
things that have changed), just being notified there is a change doesn't he=
lp with caching.
>
> Ben
>
>>
>> AFAIK, In the ALTO context, HTTP is similar to a transport mechanism. Th=
e
>> validity of the map information is related to the ALTO layer and delegat=
ing
>> this to the HTTP layer is probably not a good idea.

Why not?  Do you see cases where HTTP's notion of caching is different
than what we want in ALTO?  (Aside from the caching a part of a
response, which Ben has already commented on above..)

>> Regarding the redirect mechanism for POST which you have suggested, it
>> involves additional messaging which can be avoided by the mechanism prop=
osed
>> in the draft.

Yes - that is one aspect of the problem, but note that there isn't a
captive user waiting for the page to load and at least in the common
case the data will not be changing at the sub-second level.  Do you
think the extra round-trip is going to matter for this data?  I'm not
sure I see the rationale for reinventing HTTP's caching.

>> (3) Reading through quickly, anyways, the notification scheme seems
>> conceptually similar to that discussed in
>> http://tools.ietf.org/html/draft-sun-alto-notification-02. =A0There was
>> extensive discussion on the list for that document, so it may be worth
>> reading through that (if you haven't already, of course).
>>
>> [SK]
>> Thanks for providing reference to sun-alto-notification draft. In our
>> proposal the notification message carries the map information (only the
>> changed) unlike the referred draft where the client needs to query again=
.
>>
>> Notification framework helps in fast propagation of information changes.=
 In
>> the recent developments in the working group, alto-next also proposes se=
rver
>> notification mechanism.
>>
>> Thanks again for submitting the draft,
>> Rich
>>
>> On Tue, Feb 21, 2012 at 9:23 PM, sreekanth madhavan
>> <sreekanth.madhavan@huawei.com> wrote:
>>> Hi All,
>>>
>>> We have submitted a new draft related to Alto caching and subscription.=
 We
>> felt that the existing HTTP based caching mechanisms are not suitable fo=
r
>> caching Alto map information.
>>>
>>> We also proposed a new mechanism for subscription and notification whic=
h
>> we think will help reduce redundant traffic between alto nodes.
>>>
>>> Your comments are highly welcome.
>>>
>>> http://www.ietf.org/id/draft-alto-caching-subscription-00.txt
>>>
>>>
>>> Regards,
>>> Sreekanth Madhavan, on behalf of all authors
>>>
>>> -----Original Message-----
>>> From: internet-drafts@ietf.org [mailto:internet-drafts@ietf.org]
>>> Sent: Wednesday, February 22, 2012 10:29 AM
>>> To: sreekanth.madhavan@huawei.com
>>> Cc: sreekanth.madhavan@huawei.com; ravi.nandiraju@huawei.com
>>> Subject: New Version Notification for
>> draft-alto-caching-subscription-00.txt
>>>
>>> A new version of I-D, draft-alto-caching-subscription-00.txt has been
>> successfully submitted by Sreekanth Madhavan and posted to the IETF
>> repository.
>>>
>>> Filename: =A0 =A0 =A0 =A0draft-alto-caching-subscription
>>> Revision: =A0 =A0 =A0 =A000
>>> Title: =A0 =A0 =A0 =A0 =A0 ALTO Caching and Subscription
>>> Creation date: =A0 2012-02-22
>>> WG ID: =A0 =A0 =A0 =A0 =A0 Individual Submission
>>> Number of pages: 19
>>>
>>> Abstract:
>>> =A0 The specification of the ALTO protocol uses map based approaches
>>> =A0 assuming that the information provided is static for a longer perio=
d
>>> =A0 of time. =A0But in some cases network operators reallocate IP subne=
ts
>>> =A0 from time to time which in turn changes the mapping partitions[I-
>>> =A0 D.ietf-alto-deployments]. =A0Since the ALTO clients are unaware of =
the
>>> =A0 map information changes, clients need to query the servers for ever=
y
>>> =A0 service request and many such requests are redundant because the
>>> =A0 information was not changed. =A0The purpose of this memo is to prov=
ide
>>> =A0 two mechanisms which will help the ALTO clients to be informed of t=
he
>>> =A0 map information changes. a) ALTO clients cache the map information
>>> =A0 and the servers provide the expiration time for invalidation. b) AL=
TO
>>> =A0 clients can subscribe for event notifications from the server
>>>
>>>
>>>
>>>
>>> The IETF Secretariat
>>>
>>> _______________________________________________
>>> alto mailing list
>>> alto@ietf.org
>>> https://www.ietf.org/mailman/listinfo/alto
>>
>> _______________________________________________
>> alto mailing list
>> alto@ietf.org
>> https://www.ietf.org/mailman/listinfo/alto
>

From martin.thomson@gmail.com  Wed Feb 22 08:09:53 2012
Return-Path: <martin.thomson@gmail.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 840D021F86B4 for <alto@ietfa.amsl.com>; Wed, 22 Feb 2012 08:09:53 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.719
X-Spam-Level: 
X-Spam-Status: No, score=-4.719 tagged_above=-999 required=5 tests=[AWL=-1.120, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RuVkOqz2sUYt for <alto@ietfa.amsl.com>; Wed, 22 Feb 2012 08:09:52 -0800 (PST)
Received: from mail-bk0-f44.google.com (mail-bk0-f44.google.com [209.85.214.44]) by ietfa.amsl.com (Postfix) with ESMTP id 748D421F8606 for <alto@ietf.org>; Wed, 22 Feb 2012 08:09:52 -0800 (PST)
Received: by bkuw12 with SMTP id w12so239176bku.31 for <alto@ietf.org>; Wed, 22 Feb 2012 08:09:51 -0800 (PST)
Received-SPF: pass (google.com: domain of martin.thomson@gmail.com designates 10.204.141.9 as permitted sender) client-ip=10.204.141.9; 
Authentication-Results: mr.google.com; spf=pass (google.com: domain of martin.thomson@gmail.com designates 10.204.141.9 as permitted sender) smtp.mail=martin.thomson@gmail.com; dkim=pass header.i=martin.thomson@gmail.com
Received: from mr.google.com ([10.204.141.9]) by 10.204.141.9 with SMTP id k9mr15902220bku.93.1329926991667 (num_hops = 1); Wed, 22 Feb 2012 08:09:51 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=PDC0zt9N19m0OaGD/r+JC7Ja5V37ZVF2JhojRZQPxyI=; b=hbzR9pMhDTOlQU5Jv5MFQ86kVrYK6xGEORuVgeeetBSNT+AnbcH+4U4C3bIf5LOzvT u5vk61lFGe59YEmfJkKxbZquxc6Kl2wvgpxe1s5rj/xbgqxriPGGAjS8MpTk4klJPkED wHRqzIcdGjxlZVGwCyvrgbi9zokfb5dF7QAcY=
MIME-Version: 1.0
Received: by 10.204.141.9 with SMTP id k9mr12792509bku.93.1329926991464; Wed, 22 Feb 2012 08:09:51 -0800 (PST)
Received: by 10.204.241.81 with HTTP; Wed, 22 Feb 2012 08:09:51 -0800 (PST)
In-Reply-To: <CA+cvDaZMv3v4Vp7yLrSHn7kHt+Ag_Q4kr+MkRGv8cHMqfWYT+Q@mail.gmail.com>
References: <000c01ccf122$29365370$7ba2fa50$@com> <CA+cvDaY0eqQ1S7RXXrfqPmX8SyaLVazL3hPcXQbSzBkXy639Vg@mail.gmail.com> <000301ccf170$eea45500$cbecff00$@com> <B4AFE7CF-0A03-4F22-ADB7-56A2EFC96E21@niven-jenkins.co.uk> <CA+cvDaZMv3v4Vp7yLrSHn7kHt+Ag_Q4kr+MkRGv8cHMqfWYT+Q@mail.gmail.com>
Date: Wed, 22 Feb 2012 08:09:51 -0800
Message-ID: <CABkgnnWEqCv6RHhO570Jo4WPkFK9Fy+_yxB35TvbsxEeTjYSkw@mail.gmail.com>
From: Martin Thomson <martin.thomson@gmail.com>
To: Richard Alimi <rich@velvetsea.net>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Cc: sreekanth madhavan <sreekanth.madhavan@huawei.com>, alto@ietf.org
Subject: Re: [alto] FW: New Version Notification for draft-alto-caching-subscription-00.txt
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 Feb 2012 16:09:53 -0000

On 22 February 2012 07:43, Richard Alimi <rich@velvetsea.net> wrote:
> Why not? =C2=A0Do you see cases where HTTP's notion of caching is differe=
nt
> than what we want in ALTO? =C2=A0(Aside from the caching a part of a
> response, which Ben has already commented on above..)

If you need to identify subsets of information, then it is quite
possible that you have too few resources in the design*.  A subset
that is capable of changing could conceivably be identified as an
independent resource with independent expiration and caching.  If you
want to retain the economy of a single, aggregated resource, you can
still identify portions within that resource independently.  It's
possible that you would require extensions to the data format to do so
and there would be tradeoffs in complexity and (maybe) chattiness.
But you could then build on HTTP caching.

One set of tradeoffs has already been made and that produced the
current design.  That doesn't prevent a well-justified and designed
extension from functioning.  There may be a case for optimization.
But there are probably better ways than inventing a protocol-specific
subscription extension.

--Martin

p.s. Have you looked at RFC 5989?

From sreekanth.madhavan@huawei.com  Wed Feb 22 21:51:10 2012
Return-Path: <sreekanth.madhavan@huawei.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5C4DF21F85D3 for <alto@ietfa.amsl.com>; Wed, 22 Feb 2012 21:51:10 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.316
X-Spam-Level: 
X-Spam-Status: No, score=-6.316 tagged_above=-999 required=5 tests=[AWL=0.283,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id nvOpwcP4wIB1 for <alto@ietfa.amsl.com>; Wed, 22 Feb 2012 21:51:05 -0800 (PST)
Received: from szxga01-in.huawei.com (szxga01-in.huawei.com [119.145.14.64]) by ietfa.amsl.com (Postfix) with ESMTP id 141E121F8447 for <alto@ietf.org>; Wed, 22 Feb 2012 21:50:42 -0800 (PST)
Received: from huawei.com (szxga05-in [172.24.2.49]) by szxga05-in.huawei.com (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0LZU002JU08AQQ@szxga05-in.huawei.com> for alto@ietf.org; Thu, 23 Feb 2012 13:50:34 +0800 (CST)
Received: from szxrg02-dlp.huawei.com ([172.24.2.119]) by szxga05-in.huawei.com (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0LZU00ILU0878Y@szxga05-in.huawei.com> for alto@ietf.org; Thu, 23 Feb 2012 13:50:34 +0800 (CST)
Received: from szxeml209-edg.china.huawei.com ([172.24.2.119]) by szxrg02-dlp.huawei.com (MOS 4.1.9-GA)	with ESMTP id AHI46860; Thu, 23 Feb 2012 13:49:03 +0800
Received: from SZXEML413-HUB.china.huawei.com (10.82.67.152) by szxeml209-edg.china.huawei.com (172.24.2.184) with Microsoft SMTP Server (TLS) id 14.1.323.3; Thu, 23 Feb 2012 13:48:40 +0800
Received: from blrprnc03ns (10.18.96.92) by szxeml413-hub.china.huawei.com (10.82.67.152) with Microsoft SMTP Server id 14.1.323.3; Thu, 23 Feb 2012 13:49:00 +0800
Date: Thu, 23 Feb 2012 11:18:59 +0530
From: sreekanth madhavan <sreekanth.madhavan@huawei.com>
In-reply-to: <B4AFE7CF-0A03-4F22-ADB7-56A2EFC96E21@niven-jenkins.co.uk>
X-Originating-IP: [10.18.96.92]
To: 'Ben Niven-Jenkins' <ben@niven-jenkins.co.uk>
Message-id: <000901ccf1ee$d77f81f0$867e85d0$@com>
MIME-version: 1.0
X-Mailer: Microsoft Office Outlook 12.0
Content-type: text/plain; charset=us-ascii
Content-language: en-us
Content-transfer-encoding: 7BIT
Thread-index: Aczxc9vqR16DCG5LQMqSEhG4ekyyQAAeh7MQ
X-CFilter-Loop: Reflected
References: <000c01ccf122$29365370$7ba2fa50$@com> <CA+cvDaY0eqQ1S7RXXrfqPmX8SyaLVazL3hPcXQbSzBkXy639Vg@mail.gmail.com> <000301ccf170$eea45500$cbecff00$@com> <B4AFE7CF-0A03-4F22-ADB7-56A2EFC96E21@niven-jenkins.co.uk>
Cc: alto@ietf.org
Subject: Re: [alto] FW: New Version Notification for draft-alto-caching-subscription-00.txt
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 Feb 2012 05:51:11 -0000

A notification mechanism doesn't help you with that problem though. What you
need is a mechanism to provide incremental updates (i.e. only send the
things that have changed), just being notified there is a change doesn't
help with caching.

[SK]
The notification message sends the changed map information for which the
clients have already subscribed. We have mentioned this point in the
draft(Introduction section). The mechanism was not clearly explained. We can
add this to the draft.
 
Regards,
Sreekanth

-----Original Message-----
From: Ben Niven-Jenkins [mailto:ben@niven-jenkins.co.uk] 
Sent: Wednesday, February 22, 2012 8:38 PM
To: sreekanth madhavan
Cc: 'Richard Alimi'; alto@ietf.org
Subject: Re: [alto] FW: New Version Notification for
draft-alto-caching-subscription-00.txt


On 22 Feb 2012, at 14:47, sreekanth madhavan wrote:
> From: richard.alimi@gmail.com [mailto:richard.alimi@gmail.com] On Behalf
Of
> Richard Alimi
> (2) There are ways to do caching in HTTP for POST requests (e.g.,
> server can reply with an alternate location and the client does a GET
> on that location).
> 
> 
> [SK]
> HTTP based caching mechanisms cannot be used for subset of the content. In
> most cases ISPs change part of the network map information related to some
> of the PIDs.
> In the above case, ALTO clients ideally will be interested only in the
> changed information and it does not make sense to resend the whole
> information to the clients.

A notification mechanism doesn't help you with that problem though. What you
need is a mechanism to provide incremental updates (i.e. only send the
things that have changed), just being notified there is a change doesn't
help with caching.

Ben

> 
> AFAIK, In the ALTO context, HTTP is similar to a transport mechanism. The
> validity of the map information is related to the ALTO layer and
delegating
> this to the HTTP layer is probably not a good idea.
> 
> Regarding the redirect mechanism for POST which you have suggested, it
> involves additional messaging which can be avoided by the mechanism
proposed
> in the draft.
> 
> (3) Reading through quickly, anyways, the notification scheme seems
> conceptually similar to that discussed in
> http://tools.ietf.org/html/draft-sun-alto-notification-02.  There was
> extensive discussion on the list for that document, so it may be worth
> reading through that (if you haven't already, of course).
> 
> [SK]
> Thanks for providing reference to sun-alto-notification draft. In our
> proposal the notification message carries the map information (only the
> changed) unlike the referred draft where the client needs to query again. 
> 
> Notification framework helps in fast propagation of information changes.
In
> the recent developments in the working group, alto-next also proposes
server
> notification mechanism.
> 
> Thanks again for submitting the draft,
> Rich
> 
> On Tue, Feb 21, 2012 at 9:23 PM, sreekanth madhavan
> <sreekanth.madhavan@huawei.com> wrote:
>> Hi All,
>> 
>> We have submitted a new draft related to Alto caching and subscription.
We
> felt that the existing HTTP based caching mechanisms are not suitable for
> caching Alto map information.
>> 
>> We also proposed a new mechanism for subscription and notification which
> we think will help reduce redundant traffic between alto nodes.
>> 
>> Your comments are highly welcome.
>> 
>> http://www.ietf.org/id/draft-alto-caching-subscription-00.txt
>> 
>> 
>> Regards,
>> Sreekanth Madhavan, on behalf of all authors
>> 
>> -----Original Message-----
>> From: internet-drafts@ietf.org [mailto:internet-drafts@ietf.org]
>> Sent: Wednesday, February 22, 2012 10:29 AM
>> To: sreekanth.madhavan@huawei.com
>> Cc: sreekanth.madhavan@huawei.com; ravi.nandiraju@huawei.com
>> Subject: New Version Notification for
> draft-alto-caching-subscription-00.txt
>> 
>> A new version of I-D, draft-alto-caching-subscription-00.txt has been
> successfully submitted by Sreekanth Madhavan and posted to the IETF
> repository.
>> 
>> Filename:        draft-alto-caching-subscription
>> Revision:        00
>> Title:           ALTO Caching and Subscription
>> Creation date:   2012-02-22
>> WG ID:           Individual Submission
>> Number of pages: 19
>> 
>> Abstract:
>>   The specification of the ALTO protocol uses map based approaches
>>   assuming that the information provided is static for a longer period
>>   of time.  But in some cases network operators reallocate IP subnets
>>   from time to time which in turn changes the mapping partitions[I-
>>   D.ietf-alto-deployments].  Since the ALTO clients are unaware of the
>>   map information changes, clients need to query the servers for every
>>   service request and many such requests are redundant because the
>>   information was not changed.  The purpose of this memo is to provide
>>   two mechanisms which will help the ALTO clients to be informed of the
>>   map information changes. a) ALTO clients cache the map information
>>   and the servers provide the expiration time for invalidation. b) ALTO
>>   clients can subscribe for event notifications from the server
>> 
>> 
>> 
>> 
>> The IETF Secretariat
>> 
>> _______________________________________________
>> alto mailing list
>> alto@ietf.org
>> https://www.ietf.org/mailman/listinfo/alto
> 
> _______________________________________________
> alto mailing list
> alto@ietf.org
> https://www.ietf.org/mailman/listinfo/alto


From ravi.nandiraju@huawei.com  Thu Feb 23 02:34:11 2012
Return-Path: <ravi.nandiraju@huawei.com>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 281DA21F84B2 for <alto@ietfa.amsl.com>; Thu, 23 Feb 2012 02:34:11 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -5.999
X-Spam-Level: 
X-Spam-Status: No, score=-5.999 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, J_CHICKENPOX_102=0.6, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qvOmJuaP2FO5 for <alto@ietfa.amsl.com>; Thu, 23 Feb 2012 02:34:10 -0800 (PST)
Received: from szxga01-in.huawei.com (szxga01-in.huawei.com [119.145.14.64]) by ietfa.amsl.com (Postfix) with ESMTP id E5FC921F846C for <alto@ietf.org>; Thu, 23 Feb 2012 02:34:09 -0800 (PST)
Received: from huawei.com (szxga05-in [172.24.2.49]) by szxga05-in.huawei.com (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0LZU00BCNDBLZ7@szxga05-in.huawei.com> for alto@ietf.org; Thu, 23 Feb 2012 18:33:22 +0800 (CST)
Received: from szxrg02-dlp.huawei.com ([172.24.2.119]) by szxga05-in.huawei.com (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0LZU00MWIDBLPK@szxga05-in.huawei.com> for alto@ietf.org; Thu, 23 Feb 2012 18:33:21 +0800 (CST)
Received: from szxeml201-edg.china.huawei.com ([172.24.2.119]) by szxrg02-dlp.huawei.com (MOS 4.1.9-GA)	with ESMTP id AHI84826; Thu, 23 Feb 2012 18:33:20 +0800
Received: from SZXEML424-HUB.china.huawei.com (10.82.67.163) by szxeml201-edg.china.huawei.com (172.24.2.39) with Microsoft SMTP Server (TLS) id 14.1.323.3; Thu, 23 Feb 2012 18:33:12 +0800
Received: from SZXEML523-MBX.china.huawei.com ([169.254.3.103]) by szxeml424-hub.china.huawei.com ([10.82.67.163]) with mapi id 14.01.0323.003; Thu, 23 Feb 2012 18:33:11 +0800
Date: Thu, 23 Feb 2012 10:33:10 +0000
From: Ravi nandiraju <ravi.nandiraju@huawei.com>
In-reply-to: <CA+cvDaZMv3v4Vp7yLrSHn7kHt+Ag_Q4kr+MkRGv8cHMqfWYT+Q@mail.gmail.com>
X-Originating-IP: [10.18.96.96]
To: Richard Alimi <rich@velvetsea.net>, Ben Niven-Jenkins <ben@niven-jenkins.co.uk>
Message-id: <7FA18E7BC523F04096632E46DA905D7A1CE6B68E@szxeml523-mbx.china.huawei.com>
MIME-version: 1.0
Content-type: text/plain; charset=us-ascii
Content-language: en-US
Content-transfer-encoding: quoted-printable
Accept-Language: en-US, zh-CN
Thread-topic: [alto] FW: New Version Notification for draft-alto-caching-subscription-00.txt
Thread-index: AQHM8XPcsGtc5fm1uUeLwjzZm/70bJZIiFUAgAHA3u0=
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
X-CFilter-Loop: Reflected
References: <000c01ccf122$29365370$7ba2fa50$@com> <CA+cvDaY0eqQ1S7RXXrfqPmX8SyaLVazL3hPcXQbSzBkXy639Vg@mail.gmail.com> <000301ccf170$eea45500$cbecff00$@com> <B4AFE7CF-0A03-4F22-ADB7-56A2EFC96E21@niven-jenkins.co.uk> <CA+cvDaZMv3v4Vp7yLrSHn7kHt+Ag_Q4kr+MkRGv8cHMqfWYT+Q@mail.gmail.com>
Cc: Sreekanth madhavan <sreekanth.madhavan@huawei.com>, "alto@ietf.org" <alto@ietf.org>
Subject: Re: [alto] FW: New Version Notification for	draft-alto-caching-subscription-00.txt
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 Feb 2012 10:34:11 -0000

Hi All,
   Thanks for all your comments.=20
The good thing I see is that everyone agrees to the point that caching of A=
LTO information is beneficial(at least from the replies). For caching of th=
is information, we have two options, use HTTP based mechanisms or define a =
new mechanism in ALTO. We(authors) seem to have a different opinion regardi=
ng the mechanism for caching. The rationale (as I understand) for using HTT=
P based caching mechanism is to reuse the existing HTTP caching infrastruct=
ure. AFAIK, many of the commonly used HTTP caches support only caching of G=
ET/HEAD requests. The POST request will be used in ALTO for filtered map qu=
eries and going forward I believe this will probably be more popular among =
the clients for the simple reason being that the client is not interested i=
n getting the whole map information every time. In this case, the existing =
cache infrastructure is not very useful. The POST caching mechanism of redi=
recting every request from the client to another URL, IMHO, is cumbersome a=
nd does not reuse the existing caching infrastructure. The server has to ge=
nerate a unique URL for every new ALTO filtered request and redirect it to =
this URL. The filtered map queries could be varied and this ends up into mu=
ltiple redirected URLs for each kind of filter.
IMO, the goal is to cache the ALTO information, HTTP caching mechanism is t=
he means. If the means is cumbersome, then it is better to devise the exten=
sion in the protocol itself.

-Ravi

________________________________________
From: alto-bounces@ietf.org [alto-bounces@ietf.org] on behalf of Richard Al=
imi [rich@velvetsea.net]
Sent: Wednesday, February 22, 2012 9:13 PM
To: Ben Niven-Jenkins
Cc: Sreekanth madhavan; alto@ietf.org
Subject: Re: [alto] FW: New Version Notification for    draft-alto-caching-=
subscription-00.txt

Weird - I didn't get Sreekanth's original reply, so replying to Ben's inste=
ad :)

On Wed, Feb 22, 2012 at 7:08 AM, Ben Niven-Jenkins
<ben@niven-jenkins.co.uk> wrote:
>
> On 22 Feb 2012, at 14:47, sreekanth madhavan wrote:
>> From: richard.alimi@gmail.com [mailto:richard.alimi@gmail.com] On Behalf=
 Of
>> Richard Alimi
>> (2) There are ways to do caching in HTTP for POST requests (e.g.,
>> server can reply with an alternate location and the client does a GET
>> on that location).
>>
>>
>> [SK]
>> HTTP based caching mechanisms cannot be used for subset of the content. =
In
>> most cases ISPs change part of the network map information related to so=
me
>> of the PIDs.
>> In the above case, ALTO clients ideally will be interested only in the
>> changed information and it does not make sense to resend the whole
>> information to the clients.
>
> A notification mechanism doesn't help you with that problem though. What =
you need is a mechanism to provide incremental updates (i.e. only send the =
things that have changed), just being notified there is a change doesn't he=
lp with caching.
>
> Ben
>
>>
>> AFAIK, In the ALTO context, HTTP is similar to a transport mechanism. Th=
e
>> validity of the map information is related to the ALTO layer and delegat=
ing
>> this to the HTTP layer is probably not a good idea.

Why not?  Do you see cases where HTTP's notion of caching is different
than what we want in ALTO?  (Aside from the caching a part of a
response, which Ben has already commented on above..)

>> Regarding the redirect mechanism for POST which you have suggested, it
>> involves additional messaging which can be avoided by the mechanism prop=
osed
>> in the draft.

Yes - that is one aspect of the problem, but note that there isn't a
captive user waiting for the page to load and at least in the common
case the data will not be changing at the sub-second level.  Do you
think the extra round-trip is going to matter for this data?  I'm not
sure I see the rationale for reinventing HTTP's caching.

>> (3) Reading through quickly, anyways, the notification scheme seems
>> conceptually similar to that discussed in
>> http://tools.ietf.org/html/draft-sun-alto-notification-02.  There was
>> extensive discussion on the list for that document, so it may be worth
>> reading through that (if you haven't already, of course).
>>
>> [SK]
>> Thanks for providing reference to sun-alto-notification draft. In our
>> proposal the notification message carries the map information (only the
>> changed) unlike the referred draft where the client needs to query again=
.
>>
>> Notification framework helps in fast propagation of information changes.=
 In
>> the recent developments in the working group, alto-next also proposes se=
rver
>> notification mechanism.
>>
>> Thanks again for submitting the draft,
>> Rich
>>
>> On Tue, Feb 21, 2012 at 9:23 PM, sreekanth madhavan
>> <sreekanth.madhavan@huawei.com> wrote:
>>> Hi All,
>>>
>>> We have submitted a new draft related to Alto caching and subscription.=
 We
>> felt that the existing HTTP based caching mechanisms are not suitable fo=
r
>> caching Alto map information.
>>>
>>> We also proposed a new mechanism for subscription and notification whic=
h
>> we think will help reduce redundant traffic between alto nodes.
>>>
>>> Your comments are highly welcome.
>>>
>>> http://www.ietf.org/id/draft-alto-caching-subscription-00.txt
>>>
>>>
>>> Regards,
>>> Sreekanth Madhavan, on behalf of all authors
>>>
>>> -----Original Message-----
>>> From: internet-drafts@ietf.org [mailto:internet-drafts@ietf.org]
>>> Sent: Wednesday, February 22, 2012 10:29 AM
>>> To: sreekanth.madhavan@huawei.com
>>> Cc: sreekanth.madhavan@huawei.com; ravi.nandiraju@huawei.com
>>> Subject: New Version Notification for
>> draft-alto-caching-subscription-00.txt
>>>
>>> A new version of I-D, draft-alto-caching-subscription-00.txt has been
>> successfully submitted by Sreekanth Madhavan and posted to the IETF
>> repository.
>>>
>>> Filename:        draft-alto-caching-subscription
>>> Revision:        00
>>> Title:           ALTO Caching and Subscription
>>> Creation date:   2012-02-22
>>> WG ID:           Individual Submission
>>> Number of pages: 19
>>>
>>> Abstract:
>>>   The specification of the ALTO protocol uses map based approaches
>>>   assuming that the information provided is static for a longer period
>>>   of time.  But in some cases network operators reallocate IP subnets
>>>   from time to time which in turn changes the mapping partitions[I-
>>>   D.ietf-alto-deployments].  Since the ALTO clients are unaware of the
>>>   map information changes, clients need to query the servers for every
>>>   service request and many such requests are redundant because the
>>>   information was not changed.  The purpose of this memo is to provide
>>>   two mechanisms which will help the ALTO clients to be informed of the
>>>   map information changes. a) ALTO clients cache the map information
>>>   and the servers provide the expiration time for invalidation. b) ALTO
>>>   clients can subscribe for event notifications from the server
>>>
>>>
>>>
>>>
>>> The IETF Secretariat
>>>
>>> _______________________________________________
>>> alto mailing list
>>> alto@ietf.org
>>> https://www.ietf.org/mailman/listinfo/alto
>>
>> _______________________________________________
>> alto mailing list
>> alto@ietf.org
>> https://www.ietf.org/mailman/listinfo/alto
>
_______________________________________________
alto mailing list
alto@ietf.org
https://www.ietf.org/mailman/listinfo/alto=

From ietfdbh@comcast.net  Thu Feb 23 08:09:57 2012
Return-Path: <ietfdbh@comcast.net>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F38EC21F876A for <alto@ietfa.amsl.com>; Thu, 23 Feb 2012 08:09:55 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -100.806
X-Spam-Level: 
X-Spam-Status: No, score=-100.806 tagged_above=-999 required=5 tests=[AWL=-1.093, BAYES_05=-1.11, HTML_MESSAGE=0.001, MIME_QP_LONG_LINE=1.396, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id s2ukOsDBa7Z6 for <alto@ietfa.amsl.com>; Thu, 23 Feb 2012 08:09:50 -0800 (PST)
Received: from qmta10.westchester.pa.mail.comcast.net (qmta10.westchester.pa.mail.comcast.net [76.96.62.17]) by ietfa.amsl.com (Postfix) with ESMTP id EC8F121F8740 for <alto@ietf.org>; Thu, 23 Feb 2012 08:09:49 -0800 (PST)
Received: from omta21.westchester.pa.mail.comcast.net ([76.96.62.72]) by qmta10.westchester.pa.mail.comcast.net with comcast id dT911i0041ZXKqc5AU9qZg; Thu, 23 Feb 2012 16:09:50 +0000
Received: from [192.168.1.33] ([71.233.85.150]) by omta21.westchester.pa.mail.comcast.net with comcast id dU9j1i00F3Ecudz3hU9ofv; Thu, 23 Feb 2012 16:09:50 +0000
User-Agent: Microsoft-MacOutlook/14.14.0.111121
Date: Thu, 23 Feb 2012 11:09:40 -0500
From: David Harrington <ietfdbh@comcast.net>
To: <alto@ietf.org>, <altoext@ietf.org>
Message-ID: <CB6BCEF4.14D7E%ietfdbh@comcast.net>
Thread-Topic: draft-marocco-alto-next-00
Mime-version: 1.0
Content-type: multipart/alternative; boundary="B_3412840189_27381811"
Subject: [alto] draft-marocco-alto-next-00
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 Feb 2012 16:09:58 -0000

> This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

--B_3412840189_27381811
Content-type: text/plain;
	charset="ISO-8859-1"
Content-transfer-encoding: quoted-printable

Hi,

AD-hat-off =8A

I am not very convinced this is a set of problems that need ALTO solutions.

When dealing with P2P scenarios, ALTO is important because endpoints for a
large amount of P2P are "unmanaged" - they are typically home users sharing
files with other home users. Home users typically do not use/monitor
protocols such as BGP, ISIS, SNMP, Conex, ECN. Frequently consumer equipmen=
t
doesn't make these protocols available/accessible to end-users.

The information about the network, like server load, link status, bandwidth
availability, is not something the network providers necessarily want to
share. Network operators should be concerned about sharing with anonymous
users, who might well be interested in maliciously attacking the network
environment.=20

Data centers and CDNs typically are "managed" environments, and the
file-sharing/load-balancing/congestion control protocols are for use within
the administrative domain by the operators of the data centers or CDNs (or
between "peered" environments, where there is a certain level of trust).
These environments typically have access to protocols such as SNMP and BGP,
and how the network is "tweaked" to accommodate dynamic traffics patterns i=
s
the business of the network provider, using specialty applications to adapt
the network at the lower layers. Operators and their OAM protocols monitor
traffic load and can set policies to balance the load/adjust the forwarding
rules as needed to compensate for congestion, and so on. Applications
running on end-hosts do not have enough knowledge of the complete network
traffic, and are in a bad position to make policy decisions about load
balancing across servers based on bandwidth availability or server load or
memory usage.

I understand that there is a need for communications between layer 7
applications and the underlying layer 4,3,and 2 functionality.There are
already protocols available that allow applications to inform the lower
layers of the network what type of traffic they plan to introduce to the
network, and the qualities of the service they prefer for their traffic.
Applications can already make use of some of the existing standards for thi=
s
purpose. Users probably do not have authorization to affect the policy; the=
y
can request QoS within the policies configured by the network operators.  I
do not see why, with few exceptions, the layer 7 application is better
positioned to be the policy decision point, especially for real-time
adjustments, than the OAM functionality already built into those lower
layers, and the network provider policy configurations. I also think that
real-time adjustments by ALTO don't seem called for, so a push model for
fast dynamic updates really isn't needed. If needed, existing push protocol=
s
such as SNMP notifications, driven by an ALTO-SERVER-MIB,  could serve this
purpose just fine.

I have a concern about server-to-server sharing of information. I think the
network provider can decide which servers to share information with. If
server-to-server sharing eliminates the network provider from the decision
of whom to share data with, I consider that a problem. You, of course, do
not discuss how sharing would be done in this document, so maybe that issue
could be addressed.

Some of these ideas, such as server-to-server communications, might be
covered by a re-charter for the WG. However, developing a brand-new protoco=
l
just for this purpose seems dubious when there are so many existing
protocols that can carry data between applications (which is what an alto
server is). I would expect that a better approach might be to have a server
and client co-resident, and using a (server-as-client)-to-server
communications.

I think you might have a hard time convincing the IESG to approve ALTO
extensions for data centers and CDNs that could already be handled by other
protocols in a managed environment.

My $.02
--
David Harrington




--B_3412840189_27381811
Content-type: text/html;
	charset="ISO-8859-1"
Content-transfer-encoding: quoted-printable

<html><head></head><body style=3D"word-wrap: break-word; -webkit-nbsp-mode: s=
pace; -webkit-line-break: after-white-space; color: rgb(0, 0, 0); font-size:=
 14px; "><div><div><div style=3D"font-family: Calibri, sans-serif; ">Hi,</div>=
<div style=3D"font-family: Calibri, sans-serif; "><br></div><div style=3D"font-f=
amily: Calibri, sans-serif; ">AD-hat-off &#8230;</div><div style=3D"font-famil=
y: Calibri, sans-serif; "><br></div><div style=3D"font-family: Calibri, sans-s=
erif; ">I am not very convinced this is a set of problems that need ALTO sol=
utions.&nbsp;</div><div style=3D"font-family: Calibri, sans-serif; "><br></div=
><div style=3D"font-family: Calibri, sans-serif; ">When dealing with P2P scena=
rios, ALTO is important because endpoints for a large amount of P2P are "unm=
anaged" - they are typically home users sharing files with other home users.=
 Home users typically do not use/monitor protocols such as BGP, ISIS, SNMP, =
Conex, ECN. Frequently consumer equipment doesn't make these protocols avail=
able/accessible to end-users.&nbsp;</div><div style=3D"font-family: Calibri, s=
ans-serif; "><br></div><div style=3D"font-family: Calibri, sans-serif; ">The i=
nformation about the network, like server load, link status, bandwidth avail=
ability, is not something the network providers necessarily want to share. N=
etwork operators should be concerned about sharing with anonymous users, who=
 might well be interested in maliciously attacking the network environment.&=
nbsp;</div><div style=3D"font-family: Calibri, sans-serif; "><br></div><div st=
yle=3D"font-family: Calibri, sans-serif; ">Data centers and CDNs typically are=
 "managed" environments, and the file-sharing/load-balancing/congestion cont=
rol protocols are for use within the administrative domain by the operators =
of the data centers or CDNs (or between "peered" environments, where there i=
s a certain level of trust). These environments typically have access to pro=
tocols such as SNMP and BGP, and how the network is "tweaked" to accommodate=
 dynamic traffics patterns is the business of the network provider, using sp=
ecialty applications to adapt the network at the lower layers. Operators and=
 their OAM protocols monitor traffic load and can set policies to balance th=
e load/adjust the forwarding rules as needed to compensate for congestion, a=
nd so on. Applications running on end-hosts do not have enough knowledge of =
the complete network traffic, and are in a bad position to make policy decis=
ions about load balancing across servers based on bandwidth availability or =
server load or memory usage.</div><div style=3D"font-family: Calibri, sans-ser=
if; "><br></div><div style=3D"font-family: Calibri, sans-serif; ">I understand=
 that there is a need for communications between layer 7 applications and th=
e underlying layer 4,3,and 2 functionality.There are already protocols avail=
able that allow applications to inform the lower layers of the network what =
type of traffic they plan to introduce to the network, and the qualities of =
the service they prefer for their traffic. Applications can already make use=
 of some of the existing standards for this purpose. Users probably do not h=
ave authorization to affect the policy; they can request QoS within the poli=
cies&nbsp;configured by the network operators.&nbsp;&nbsp;I do not see why, =
with few exceptions, the layer 7 application is better positioned to be the =
policy decision point, especially for real-time adjustments, than the OAM fu=
nctionality already built into those lower layers, and the network provider =
policy configurations. I also think that real-time adjustments by ALTO don't=
 seem called for, so a push model for fast dynamic updates really isn't need=
ed. If needed, existing push protocols such as SNMP notifications, driven by=
 an ALTO-SERVER-MIB, &nbsp;could serve this purpose just fine.</div><div sty=
le=3D"font-family: Calibri, sans-serif; "><br></div><div style=3D"font-family: C=
alibri, sans-serif; ">I have a concern about server-to-server sharing of inf=
ormation. I think the network provider can decide which servers to share inf=
ormation with. If server-to-server sharing eliminates the network provider f=
rom the decision of whom to share data with, I consider that a problem. You,=
 of course, do not discuss how sharing would be done in this document, so ma=
ybe that issue could be addressed.&nbsp;</div><div style=3D"font-family: Calib=
ri, sans-serif; "><br></div><div style=3D"font-family: Calibri, sans-serif; ">=
Some of these ideas, such as server-to-server communications, might be cover=
ed by a re-charter for the WG. However, developing a brand-new protocol just=
 for this purpose seems dubious when there are so many existing protocols th=
at can carry data between applications (which is what an alto server is). I =
would expect that a better approach might be to have a server and client co-=
resident, and using a (server-as-client)-to-server communications.</div><div=
 style=3D"font-family: Calibri, sans-serif; "><br></div><div style=3D"font-famil=
y: Calibri, sans-serif; ">I think you might have a hard time convincing the =
IESG to approve ALTO extensions for data centers and CDNs that could already=
 be handled by other protocols in a managed environment.</div><div style=3D"fo=
nt-family: Calibri, sans-serif; "><br></div><div style=3D"font-family: Calibri=
, sans-serif; ">My $.02</div><div><div style=3D"font-family: Calibri, sans-ser=
if; "><font class=3D"Apple-style-span" color=3D"rgb(0, 0, 0)"><font class=3D"Apple=
-style-span" face=3D"Calibri">--</font></font></div><div style=3D"font-family: C=
alibri, sans-serif; "><font class=3D"Apple-style-span" color=3D"rgb(0, 0, 0)"><f=
ont class=3D"Apple-style-span" face=3D"Calibri"><span class=3D"Apple-style-span" s=
tyle=3D"font-size: 14px;">David Harrington</span></font></font></div><div styl=
e=3D"font-family: Calibri, sans-serif; "><font class=3D"Apple-style-span" color=3D=
"rgb(0, 0, 0)"><font class=3D"Apple-style-span" face=3D"Calibri"><span class=3D"Ap=
ple-style-span" style=3D"font-size: 14px;"><br></span></font></font></div></di=
v></div></div></body></html>

--B_3412840189_27381811--



From enrico.marocco@telecomitalia.it  Mon Feb 27 01:03:51 2012
Return-Path: <enrico.marocco@telecomitalia.it>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2D6B321F8562; Mon, 27 Feb 2012 01:03:51 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -99.637
X-Spam-Level: 
X-Spam-Status: No, score=-99.637 tagged_above=-999 required=5 tests=[AWL=-0.777, BAYES_20=-0.74, HELO_EQ_IT=0.635, HOST_EQ_IT=1.245,  USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 8WXsjPUhHWAN; Mon, 27 Feb 2012 01:03:50 -0800 (PST)
Received: from GRFEDG701BA020.telecomitalia.it (grfedg701ba020.telecomitalia.it [156.54.233.200]) by ietfa.amsl.com (Postfix) with ESMTP id CADC421F855A; Mon, 27 Feb 2012 01:03:49 -0800 (PST)
Received: from GRFHUB701BA020.griffon.local (10.188.101.111) by GRFEDG701BA020.telecomitalia.it (10.188.45.100) with Microsoft SMTP Server (TLS) id 8.2.254.0; Mon, 27 Feb 2012 10:03:41 +0100
Received: from MacLab.local (163.162.180.246) by smtp.telecomitalia.it (10.188.101.114) with Microsoft SMTP Server (TLS) id 8.2.254.0; Mon, 27 Feb 2012 10:03:40 +0100
Message-ID: <4F4B4756.7060603@telecomitalia.it>
Date: Mon, 27 Feb 2012 10:05:26 +0100
From: Enrico Marocco <enrico.marocco@telecomitalia.it>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:10.0.2) Gecko/20120216 Thunderbird/10.0.2
MIME-Version: 1.0
To: David Harrington <ietfdbh@comcast.net>
References: <CB6BCEF4.14D7E%ietfdbh@comcast.net>
In-Reply-To: <CB6BCEF4.14D7E%ietfdbh@comcast.net>
Content-Type: multipart/signed; protocol="application/pkcs7-signature"; micalg=sha1; boundary="------------ms000404090604030505070508"
X-TI-Disclaimer: Disclaimer1
Cc: "altoext@ietf.org" <altoext@ietf.org>, "alto@ietf.org" <alto@ietf.org>
Subject: Re: [alto] [altoext] draft-marocco-alto-next-00
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 27 Feb 2012 09:03:51 -0000

--------------ms000404090604030505070508
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Thanks Dave,

this is indeed the kind of feedback the draft was intended to trigger.

Proposals that have been presented in the past few years for extending
the ALTO protocol on different axis, to make it suitable for carrying
different types of information and for doing that on a more dynamic
timescale, seem to indicate that there is some sort of gap in the
provisioning of network information to applications that tools currently
available have failed to fill. If that is the case, the discussion we
want to have needs to deepen in the whys and hows those tools do not
satisfy requirements of such applications -- I'm certainly not the most
expert in the area and will leave the question to the proponents of the
extensions, but I guess the answer may be related to the intrinsic
complexity of dealing with low-level information at an application level
-- and why a more abstract presentation of such information (through
ALTO or via other means) may have better chance.

You make a good point that for CDN and datacenter applications running
in controlled environment SNMP monitoring and BGP sniffing can provide
all the information such applications may possibly need. If the
incentive they provide is enough to motivate their implementation is a
key topic of discussion we want to have to identify the way forward for
ALTO.

Enrico

On 2/23/12 5:09 PM, David Harrington wrote:
> Hi,
>=20
> AD-hat-off =85
>=20
> I am not very convinced this is a set of problems that need ALTO soluti=
ons.=20
>=20
> When dealing with P2P scenarios, ALTO is important because endpoints fo=
r
> a large amount of P2P are "unmanaged" - they are typically home users
> sharing files with other home users. Home users typically do not
> use/monitor protocols such as BGP, ISIS, SNMP, Conex, ECN. Frequently
> consumer equipment doesn't make these protocols available/accessible to=

> end-users.=20
>=20
> The information about the network, like server load, link status,
> bandwidth availability, is not something the network providers
> necessarily want to share. Network operators should be concerned about
> sharing with anonymous users, who might well be interested in
> maliciously attacking the network environment.=20
>=20
> Data centers and CDNs typically are "managed" environments, and the
> file-sharing/load-balancing/congestion control protocols are for use
> within the administrative domain by the operators of the data centers o=
r
> CDNs (or between "peered" environments, where there is a certain level
> of trust). These environments typically have access to protocols such a=
s
> SNMP and BGP, and how the network is "tweaked" to accommodate dynamic
> traffics patterns is the business of the network provider, using
> specialty applications to adapt the network at the lower layers.
> Operators and their OAM protocols monitor traffic load and can set
> policies to balance the load/adjust the forwarding rules as needed to
> compensate for congestion, and so on. Applications running on end-hosts=

> do not have enough knowledge of the complete network traffic, and are i=
n
> a bad position to make policy decisions about load balancing across
> servers based on bandwidth availability or server load or memory usage.=

>=20
> I understand that there is a need for communications between layer 7
> applications and the underlying layer 4,3,and 2 functionality.There are=

> already protocols available that allow applications to inform the lower=

> layers of the network what type of traffic they plan to introduce to th=
e
> network, and the qualities of the service they prefer for their traffic=
=2E
> Applications can already make use of some of the existing standards for=

> this purpose. Users probably do not have authorization to affect the
> policy; they can request QoS within the policies configured by the
> network operators.  I do not see why, with few exceptions, the layer 7
> application is better positioned to be the policy decision point,
> especially for real-time adjustments, than the OAM functionality alread=
y
> built into those lower layers, and the network provider policy
> configurations. I also think that real-time adjustments by ALTO don't
> seem called for, so a push model for fast dynamic updates really isn't
> needed. If needed, existing push protocols such as SNMP notifications,
> driven by an ALTO-SERVER-MIB,  could serve this purpose just fine.
>=20
> I have a concern about server-to-server sharing of information. I think=

> the network provider can decide which servers to share information with=
=2E
> If server-to-server sharing eliminates the network provider from the
> decision of whom to share data with, I consider that a problem. You, of=

> course, do not discuss how sharing would be done in this document, so
> maybe that issue could be addressed.=20
>=20
> Some of these ideas, such as server-to-server communications, might be
> covered by a re-charter for the WG. However, developing a brand-new
> protocol just for this purpose seems dubious when there are so many
> existing protocols that can carry data between applications (which is
> what an alto server is). I would expect that a better approach might be=

> to have a server and client co-resident, and using a
> (server-as-client)-to-server communications.
>=20
> I think you might have a hard time convincing the IESG to approve ALTO
> extensions for data centers and CDNs that could already be handled by
> other protocols in a managed environment.
>=20
> My $.02
> --
> David Harrington
>=20



--------------ms000404090604030505070508
Content-Type: application/pkcs7-signature; name="smime.p7s"
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="smime.p7s"
Content-Description: S/MIME Cryptographic Signature
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--------------ms000404090604030505070508--

From yry@cs.yale.edu  Wed Feb 29 20:54:15 2012
Return-Path: <yry@cs.yale.edu>
X-Original-To: alto@ietfa.amsl.com
Delivered-To: alto@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 100D321F8562 for <alto@ietfa.amsl.com>; Wed, 29 Feb 2012 20:54:15 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.598
X-Spam-Level: 
X-Spam-Status: No, score=-2.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 7OD+L28ykBTM for <alto@ietfa.amsl.com>; Wed, 29 Feb 2012 20:54:11 -0800 (PST)
Received: from vm-emlprdomr-05.its.yale.edu (vm-emlprdomr-05.its.yale.edu [130.132.50.146]) by ietfa.amsl.com (Postfix) with ESMTP id 69CA421E801E for <alto@ietf.org>; Wed, 29 Feb 2012 20:54:11 -0800 (PST)
Received: from dhcp-128-36-168-225.central.yale.edu (dhcp-128-36-168-225.central.yale.edu [128.36.168.225]) (authenticated bits=0) by vm-emlprdomr-05.its.yale.edu (8.14.4/8.14.4) with ESMTP id q214s92a016703 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Wed, 29 Feb 2012 23:54:09 -0500
Message-ID: <4F4F00F1.9070108@cs.yale.edu>
Date: Wed, 29 Feb 2012 23:54:09 -0500
From: "Y. Richard Yang" <yry@cs.yale.edu>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:10.0.2) Gecko/20120216 Thunderbird/10.0.2
MIME-Version: 1.0
To: alto@ietf.org
References: <CB6BCEF4.14D7E%ietfdbh@comcast.net> <4F4B4756.7060603@telecomitalia.it>
In-Reply-To: <4F4B4756.7060603@telecomitalia.it>
Content-Type: multipart/alternative; boundary="------------000409000505040503080805"
X-Scanned-By: MIMEDefang 2.71 on 130.132.50.146
Subject: Re: [alto] [altoext] draft-marocco-alto-next-00
X-BeenThere: alto@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Application-Layer Traffic Optimization \(alto\) WG mailing list" <alto.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/alto>, <mailto:alto-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/alto>
List-Post: <mailto:alto@ietf.org>
List-Help: <mailto:alto-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/alto>, <mailto:alto-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 01 Mar 2012 04:54:15 -0000

This is a multi-part message in MIME format.
--------------000409000505040503080805
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Enrico and Dave,

Good discussions. Here is my $.01 input on data centers/controlled 
environments.
Two examples came to mind:

- A datacenter may consist of multiple frameworks to schedule 
applications. A
popular example is the Map-Reduce (Hadoop) framework. Preliminary network
aware capabilities exist; for example, see Rack Awareness
(http://hadoop.apache.org/common/docs/current/hdfs_user_guide.html#Rack+Awareness).
The document 
(https://issues.apache.org/jira/secure/attachment/12345251/Rack_aware_HDFS_proposal.pdf) 

discusses distances among nodes in simple settings as well.

- In 
(http://static.usenix.org/events/hotice11/tech/full_papers/Wang_Ye.pdf),
motivated by the need of MS Lync, the framework aggregates information 
beyond
network-centric but also app level information to provide an enterprise 
information
portal.

For both settings, I feel that more abstract and more aggregated 
presentation of
information, say using ALTO, can be valuable/preferred.

Richard

On 2/27/12 4:05 AM, Enrico Marocco wrote:
> Thanks Dave,
>
> this is indeed the kind of feedback the draft was intended to trigger.
>
> Proposals that have been presented in the past few years for extending
> the ALTO protocol on different axis, to make it suitable for carrying
> different types of information and for doing that on a more dynamic
> timescale, seem to indicate that there is some sort of gap in the
> provisioning of network information to applications that tools currently
> available have failed to fill. If that is the case, the discussion we
> want to have needs to deepen in the whys and hows those tools do not
> satisfy requirements of such applications -- I'm certainly not the most
> expert in the area and will leave the question to the proponents of the
> extensions, but I guess the answer may be related to the intrinsic
> complexity of dealing with low-level information at an application level
> -- and why a more abstract presentation of such information (through
> ALTO or via other means) may have better chance.
>
> You make a good point that for CDN and datacenter applications running
> in controlled environment SNMP monitoring and BGP sniffing can provide
> all the information such applications may possibly need. If the
> incentive they provide is enough to motivate their implementation is a
> key topic of discussion we want to have to identify the way forward for
> ALTO.
>
> Enrico
>
> On 2/23/12 5:09 PM, David Harrington wrote:
>> Hi,
>>
>> AD-hat-off ...
>>
>> I am not very convinced this is a set of problems that need ALTO solutions.
>>
>> When dealing with P2P scenarios, ALTO is important because endpoints for
>> a large amount of P2P are "unmanaged" - they are typically home users
>> sharing files with other home users. Home users typically do not
>> use/monitor protocols such as BGP, ISIS, SNMP, Conex, ECN. Frequently
>> consumer equipment doesn't make these protocols available/accessible to
>> end-users.
>>
>> The information about the network, like server load, link status,
>> bandwidth availability, is not something the network providers
>> necessarily want to share. Network operators should be concerned about
>> sharing with anonymous users, who might well be interested in
>> maliciously attacking the network environment.
>>
>> Data centers and CDNs typically are "managed" environments, and the
>> file-sharing/load-balancing/congestion control protocols are for use
>> within the administrative domain by the operators of the data centers or
>> CDNs (or between "peered" environments, where there is a certain level
>> of trust). These environments typically have access to protocols such as
>> SNMP and BGP, and how the network is "tweaked" to accommodate dynamic
>> traffics patterns is the business of the network provider, using
>> specialty applications to adapt the network at the lower layers.
>> Operators and their OAM protocols monitor traffic load and can set
>> policies to balance the load/adjust the forwarding rules as needed to
>> compensate for congestion, and so on. Applications running on end-hosts
>> do not have enough knowledge of the complete network traffic, and are in
>> a bad position to make policy decisions about load balancing across
>> servers based on bandwidth availability or server load or memory usage.
>>
>> I understand that there is a need for communications between layer 7
>> applications and the underlying layer 4,3,and 2 functionality.There are
>> already protocols available that allow applications to inform the lower
>> layers of the network what type of traffic they plan to introduce to the
>> network, and the qualities of the service they prefer for their traffic.
>> Applications can already make use of some of the existing standards for
>> this purpose. Users probably do not have authorization to affect the
>> policy; they can request QoS within the policies configured by the
>> network operators.  I do not see why, with few exceptions, the layer 7
>> application is better positioned to be the policy decision point,
>> especially for real-time adjustments, than the OAM functionality already
>> built into those lower layers, and the network provider policy
>> configurations. I also think that real-time adjustments by ALTO don't
>> seem called for, so a push model for fast dynamic updates really isn't
>> needed. If needed, existing push protocols such as SNMP notifications,
>> driven by an ALTO-SERVER-MIB,  could serve this purpose just fine.
>>
>> I have a concern about server-to-server sharing of information. I think
>> the network provider can decide which servers to share information with.
>> If server-to-server sharing eliminates the network provider from the
>> decision of whom to share data with, I consider that a problem. You, of
>> course, do not discuss how sharing would be done in this document, so
>> maybe that issue could be addressed.
>>
>> Some of these ideas, such as server-to-server communications, might be
>> covered by a re-charter for the WG. However, developing a brand-new
>> protocol just for this purpose seems dubious when there are so many
>> existing protocols that can carry data between applications (which is
>> what an alto server is). I would expect that a better approach might be
>> to have a server and client co-resident, and using a
>> (server-as-client)-to-server communications.
>>
>> I think you might have a hard time convincing the IESG to approve ALTO
>> extensions for data centers and CDNs that could already be handled by
>> other protocols in a managed environment.
>>
>> My $.02
>> --
>> David Harrington
>>
>
>
>
> _______________________________________________
> alto mailing list
> alto@ietf.org
> https://www.ietf.org/mailman/listinfo/alto


--------------000409000505040503080805
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">
    Enrico and Dave,<br>
    <br>
    Good discussions. Here is my $.01 input on data centers/controlled
    environments. <br>
    Two examples came to mind:<br>
    <br>
    - A datacenter may consist of multiple frameworks to schedule
    applications. A<br>
    popular example is the Map-Reduce (Hadoop) framework. Preliminary
    network<br>
    aware capabilities exist; for example, see Rack Awareness<br>
(<a class="moz-txt-link-freetext" href="http://hadoop.apache.org/common/docs/current/hdfs_user_guide.html#Rack+Awareness">http://hadoop.apache.org/common/docs/current/hdfs_user_guide.html#Rack+Awareness</a>).<br>
    The document
    (<a class="moz-txt-link-freetext" href="https://issues.apache.org/jira/secure/attachment/12345251/Rack_aware_HDFS_proposal.pdf">https://issues.apache.org/jira/secure/attachment/12345251/Rack_aware_HDFS_proposal.pdf</a>)
    <br>
    discusses distances among nodes in simple settings as well. <br>
    <br>
    - In
    (<a class="moz-txt-link-freetext" href="http://static.usenix.org/events/hotice11/tech/full_papers/Wang_Ye.pdf">http://static.usenix.org/events/hotice11/tech/full_papers/Wang_Ye.pdf</a>),
    <br>
    motivated by the need of MS Lync, the framework aggregates
    information beyond <br>
    network-centric but also app level information to provide an
    enterprise information <br>
    portal. <br>
    <br>
    For both settings, I feel that more abstract and more aggregated
    presentation of <br>
    information, say using ALTO, can be valuable/preferred.<br>
    <br>
    Richard<br>
    <br>
    On 2/27/12 4:05 AM, Enrico Marocco wrote:
    <blockquote cite="mid:4F4B4756.7060603@telecomitalia.it" type="cite">
      <pre wrap="">Thanks Dave,

this is indeed the kind of feedback the draft was intended to trigger.

Proposals that have been presented in the past few years for extending
the ALTO protocol on different axis, to make it suitable for carrying
different types of information and for doing that on a more dynamic
timescale, seem to indicate that there is some sort of gap in the
provisioning of network information to applications that tools currently
available have failed to fill. If that is the case, the discussion we
want to have needs to deepen in the whys and hows those tools do not
satisfy requirements of such applications -- I'm certainly not the most
expert in the area and will leave the question to the proponents of the
extensions, but I guess the answer may be related to the intrinsic
complexity of dealing with low-level information at an application level
-- and why a more abstract presentation of such information (through
ALTO or via other means) may have better chance.

You make a good point that for CDN and datacenter applications running
in controlled environment SNMP monitoring and BGP sniffing can provide
all the information such applications may possibly need. If the
incentive they provide is enough to motivate their implementation is a
key topic of discussion we want to have to identify the way forward for
ALTO.

Enrico

On 2/23/12 5:09 PM, David Harrington wrote:
</pre>
      <blockquote type="cite">
        <pre wrap="">Hi,

AD-hat-off &#8230;

I am not very convinced this is a set of problems that need ALTO solutions. 

When dealing with P2P scenarios, ALTO is important because endpoints for
a large amount of P2P are "unmanaged" - they are typically home users
sharing files with other home users. Home users typically do not
use/monitor protocols such as BGP, ISIS, SNMP, Conex, ECN. Frequently
consumer equipment doesn't make these protocols available/accessible to
end-users. 

The information about the network, like server load, link status,
bandwidth availability, is not something the network providers
necessarily want to share. Network operators should be concerned about
sharing with anonymous users, who might well be interested in
maliciously attacking the network environment. 

Data centers and CDNs typically are "managed" environments, and the
file-sharing/load-balancing/congestion control protocols are for use
within the administrative domain by the operators of the data centers or
CDNs (or between "peered" environments, where there is a certain level
of trust). These environments typically have access to protocols such as
SNMP and BGP, and how the network is "tweaked" to accommodate dynamic
traffics patterns is the business of the network provider, using
specialty applications to adapt the network at the lower layers.
Operators and their OAM protocols monitor traffic load and can set
policies to balance the load/adjust the forwarding rules as needed to
compensate for congestion, and so on. Applications running on end-hosts
do not have enough knowledge of the complete network traffic, and are in
a bad position to make policy decisions about load balancing across
servers based on bandwidth availability or server load or memory usage.

I understand that there is a need for communications between layer 7
applications and the underlying layer 4,3,and 2 functionality.There are
already protocols available that allow applications to inform the lower
layers of the network what type of traffic they plan to introduce to the
network, and the qualities of the service they prefer for their traffic.
Applications can already make use of some of the existing standards for
this purpose. Users probably do not have authorization to affect the
policy; they can request QoS within the policies configured by the
network operators.  I do not see why, with few exceptions, the layer 7
application is better positioned to be the policy decision point,
especially for real-time adjustments, than the OAM functionality already
built into those lower layers, and the network provider policy
configurations. I also think that real-time adjustments by ALTO don't
seem called for, so a push model for fast dynamic updates really isn't
needed. If needed, existing push protocols such as SNMP notifications,
driven by an ALTO-SERVER-MIB,  could serve this purpose just fine.

I have a concern about server-to-server sharing of information. I think
the network provider can decide which servers to share information with.
If server-to-server sharing eliminates the network provider from the
decision of whom to share data with, I consider that a problem. You, of
course, do not discuss how sharing would be done in this document, so
maybe that issue could be addressed. 

Some of these ideas, such as server-to-server communications, might be
covered by a re-charter for the WG. However, developing a brand-new
protocol just for this purpose seems dubious when there are so many
existing protocols that can carry data between applications (which is
what an alto server is). I would expect that a better approach might be
to have a server and client co-resident, and using a
(server-as-client)-to-server communications.

I think you might have a hard time convincing the IESG to approve ALTO
extensions for data centers and CDNs that could already be handled by
other protocols in a managed environment.

My $.02
--
David Harrington

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

</pre>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
alto mailing list
<a class="moz-txt-link-abbreviated" href="mailto:alto@ietf.org">alto@ietf.org</a>
<a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/alto">https://www.ietf.org/mailman/listinfo/alto</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>

--------------000409000505040503080805--
