From razvan@nict.go.jp Sun Jan 15 17:06:17 2012 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F332721F8474 for ; Sun, 15 Jan 2012 17:06:16 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: 1.245 X-Spam-Level: * X-Spam-Status: No, score=1.245 tagged_above=-999 required=5 tests=[BAYES_50=0.001, HELO_EQ_JP=1.244] 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 goRdxXEcmWpN for ; Sun, 15 Jan 2012 17:06:16 -0800 (PST) Received: from ns1.nict.go.jp (ns1.nict.go.jp [IPv6:2001:df0:232:300::1]) by ietfa.amsl.com (Postfix) with ESMTP id 354F921F8472 for ; Sun, 15 Jan 2012 17:06:15 -0800 (PST) Received: from gw1.nict.go.jp (gw1 [133.243.18.250]) by ns1.nict.go.jp with ESMTP id q0G16Atp012715 for ; Mon, 16 Jan 2012 10:06:10 +0900 (JST) Received: from gw1.nict.go.jp (localhost [127.0.0.1]) by gw1.nict.go.jp with ESMTP id q0G16Amn014561 for ; Mon, 16 Jan 2012 10:06:10 +0900 (JST) Received: from mail3.nict.go.jp (mail.nict.go.jp [133.243.18.3]) by gw1.nict.go.jp with ESMTP id q0G16AY0014558 for ; Mon, 16 Jan 2012 10:06:10 +0900 (JST) Received: from mail3.nict.go.jp (localhost [127.0.0.1]) by mail3.nict.go.jp (NICT Mail) with ESMTP id 03C96164F5 for ; Mon, 16 Jan 2012 10:06:10 +0900 (JST) Received: from ssh (ssh.nict.go.jp [133.243.3.49]) by mail3.nict.go.jp (NICT Mail) with ESMTP id D89B3164D8 for ; Mon, 16 Jan 2012 10:06:09 +0900 (JST) Date: Mon, 16 Jan 2012 10:06:09 +0900 (JST) From: Razvan BEURAN To: dtn-users@irtf.org Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Subject: [dtn-users] dtnping performance issues X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 16 Jan 2012 01:10:26 -0000 Dear all, I have started using DTN2 several months ago, and it has generally been a positive experience. However, some of the experiments I did produced some unexpected results in terms of performance, and I became even more puzzled after doing similar experiments with another DTN implementation most of you may already know, IBR-DTN. Since it would be too lengthy to dicuss all the problems I had, I shall focus here on a single and very simple issue, the RTT shown by dtnping when run between two PCs: * DTN2 dtnping (temporary storage in file, persistent storage in BerkleyDB): 90 ms * DTN2 dtnping (temporary storage in file, persistent storage in memory): 41 ms * IBR-DTN dtnping (temporary storage in file, persistent storage in file): 4.2 ms * IBR-DTN dtnping (temporary storage in file, persistent storage in memory): 3.6 ms * IBR-DTN dtnping (temporary storage in memory, persistent storage in memory): 2.9 ms FYI, the PCs are relatively old but fast enough (Dual-core Pentium 4, 3.2 GHz, 4GB RAM) and are connected via a switch. The standard ping command shows an RTT of about 1 ms between these 2 PCs. The results for DTN2 get even worse when using a large number of nodes (10 and 26), but since these large-experiment conditions are difficult to reproduce I will skip this issue for the time being. Given the above, I have two (sets of) questions: 1) Are the DTN2 dtnping results I obtained normal? Can anyone provide results from a similar experiment for comparison purposes. Has anyone done similar experiments with IBR-DTN? Are there any suggestions for improving performance of DTN2? 2) Has anyone done DTN2 experiments with a large number of nodes? By large I mean at least more than 10, but preferably more than 25. My preliminary results with dtnping show quite a poor performance of DTN2 compared with IBR-DTN in large settings, and I can provide more details about this if requested. Best wishes, Razvan From william.d.ivancic@nasa.gov Tue Jan 17 06:13:38 2012 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6A13D21F856F for ; Tue, 17 Jan 2012 06:13:38 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.414 X-Spam-Level: X-Spam-Status: No, score=-4.414 tagged_above=-999 required=5 tests=[AWL=-1.300, BAYES_50=0.001, HTML_FONT_FACE_BAD=0.884, 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 80OZKPyJyoT3 for ; Tue, 17 Jan 2012 06:13:37 -0800 (PST) Received: from ndjsnpf02.ndc.nasa.gov (ndjsnpf02.ndc.nasa.gov [198.117.1.122]) by ietfa.amsl.com (Postfix) with ESMTP id 961EE21F8557 for ; Tue, 17 Jan 2012 06:13:37 -0800 (PST) Received: from ndjsppt04.ndc.nasa.gov (ndjsppt04.ndc.nasa.gov [198.117.1.103]) by ndjsnpf02.ndc.nasa.gov (Postfix) with ESMTP id 05D7EA8472; Tue, 17 Jan 2012 08:13:36 -0600 (CST) Received: from ndjshub01.ndc.nasa.gov (ndjshub01-pub.ndc.nasa.gov [198.117.1.160]) by ndjsppt04.ndc.nasa.gov (8.14.4/8.14.4) with ESMTP id q0HEDZTP002573; Tue, 17 Jan 2012 08:13:35 -0600 Received: from NDJSSCC07.ndc.nasa.gov ([198.117.4.178]) by ndjshub01.ndc.nasa.gov ([198.117.1.160]) with mapi; Tue, 17 Jan 2012 08:13:35 -0600 From: "Ivancic, William D. (GRC-RHN0)" To: Razvan BEURAN Date: Tue, 17 Jan 2012 08:13:34 -0600 Thread-Topic: [dtn-users] dtnping performance issues Thread-Index: AczVIjMp44GWnEKIR8GXrY41F1hilA== Message-ID: <2881AA48-9DAA-4B60-8D9C-C50F02E25112@nasa.gov> References: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: multipart/alternative; boundary="_000_2881AA489DAA4B608D9CC50F02E25112nasagov_" MIME-Version: 1.0 X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.5.7110, 1.0.211, 0.0.0000 definitions=2012-01-17_05:2012-01-17, 2012-01-17, 1970-01-01 signatures=0 Cc: "dtn-users@irtf.org" Subject: Re: [dtn-users] dtnping performance issues X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 17 Jan 2012 14:13:38 -0000 --_000_2881AA489DAA4B608D9CC50F02E25112nasagov_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Just for clarification, I assume you are using TCP convergence layer and al= l systems have continuous connectivity? Is that correct? I don't know if either of these will matter, but for clarification: Are you using any routing protocol or static routes? Are you using DNS lookups? - Will On Jan 15, 2012, at 8:06 PM, Razvan BEURAN wrote: Dear all, I have started using DTN2 several months ago, and it has generally been a positive experience. However, some of the experiments I did produced some unexpected results in terms of performance, and I became even more puzzled after doing similar experiments with another DTN implementation most of you may already know, IBR-DTN. Since it would be too lengthy to dicuss all the problems I had, I shall focus here on a single and very simple issue, the RTT shown by dtnping when run between two PCs: * DTN2 dtnping (temporary storage in file, persistent storage in BerkleyDB): 90 ms * DTN2 dtnping (temporary storage in file, persistent storage in memory): 41 ms * IBR-DTN dtnping (temporary storage in file, persistent storage in file): 4.2 ms * IBR-DTN dtnping (temporary storage in file, persistent storage in memory): 3.6 ms * IBR-DTN dtnping (temporary storage in memory, persistent storage in memory): 2.9 ms FYI, the PCs are relatively old but fast enough (Dual-core Pentium 4, 3.2 GHz, 4GB RAM) and are connected via a switch. The standard ping command shows an RTT of about 1 ms between these 2 PCs. The results for DTN2 get even worse when using a large number of nodes (10 and 26), but since these large-experiment conditions are difficult to reproduce I will skip this issue for the time being. Given the above, I have two (sets of) questions: 1) Are the DTN2 dtnping results I obtained normal? Can anyone provide results from a similar experiment for comparison purposes. Has anyone done similar experiments with IBR-DTN? Are there any suggestions for improving performance of DTN2? 2) Has anyone done DTN2 experiments with a large number of nodes? By large I mean at least more than 10, but preferably more than 25. My preliminary results with dtnping show quite a poor performance of DTN2 compared with IBR-DTN in large settings, and I can provide more details about this if requested. Best wishes, Razvan _______________________________________________ dtn-users mailing list dtn-users@irtf.org https://www.irtf.org/mailman/listinfo/dtn-users --_000_2881AA489DAA4B608D9CC50F02E25112nasagov_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Just for clarification, I = assume you are using TCP convergence layer and all systems have continuous = connectivity?  Is that correct?

I don't know if eit= her of these will matter, but for clarification:

A= re you using any routing protocol or static routes?

Are you using DNS lookups?

- Will


On Jan 15, 2012, at 8:06 PM, Razvan BEURAN wrote:
<= br>Dear all,

I have started using DTN2 several months ago, and it ha= s generally been a
positive experience. However, some of the experiment= s I did produced some
unexpected results in terms of performance, and I= became even more puzzled
after doing similar experiments with another = DTN implementation most of
you may already know, IBR-DTN.

Since = it would be too lengthy to dicuss all the problems I had, I shall
focus= here on a single and very simple issue, the RTT shown by dtnping
when = run between two PCs:

* DTN2 dtnping (temporary storage in file,
=   persistent storage in BerkleyDB): 90 ms
* DTN2 dtnping (temporary storage in file,
&nbs= p; persistent storage in memory): 41 ms
* IBR-DTN dtnping (temporary storage in file,
 &= nbsp;persistent storage in file):        = ;            &n= bsp;     4.2 ms
* IBR-DTN dtnping (temporary st= orage in file,
  persistent storage in memory): 3.6 ms
* IBR-DTN dtnping (temporary st= orage in memory,
  persistent storage in memory): 2.9 ms

FYI, the PCs are relatively= old but fast enough (Dual-core Pentium 4, 3.2
GHz, 4GB RAM) and are co= nnected via a switch. The standard ping command
shows an RTT of about 1= ms between these 2 PCs.

The results for DTN2 get even worse when us= ing a large number of nodes (10
and 26), but since these large-experime= nt conditions are difficult to
reproduce I will skip this issue for the= time being.

Given the above, I have two (sets of) questions:
1) Are the DTN2 dtnping results I obtained normal? Can anyone provide
=    results from a similar experiment for comparison purposes= . Has anyone
   done similar experiments with IBR-DTN? A= re there any suggestions for
   improving performance of= DTN2?

2) Has anyone done DTN2 experiments with a large number of no= des? By large
   I mean at least more than 10, but prefe= rably more than 25. My
   preliminary results with dtnpi= ng show quite a poor performance of DTN2
   compared wit= h IBR-DTN in large settings, and I can provide more details
  = ; about this if requested.

Best wishes,
Razvan
__________= _____________________________________
dtn-users mailing list
dtn-users@irtf.org
https://www.irtf.o= rg/mailman/listinfo/dtn-users


= --_000_2881AA489DAA4B608D9CC50F02E25112nasagov_-- From jzinky@bbn.com Tue Jan 17 06:29:36 2012 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EEA3021F86D6 for ; Tue, 17 Jan 2012 06:29:35 -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 lyLlWIU3ycFV for ; Tue, 17 Jan 2012 06:29:35 -0800 (PST) Received: from smtp.bbn.com (smtp.bbn.com [128.33.1.81]) by ietfa.amsl.com (Postfix) with ESMTP id 0C82221F86C4 for ; Tue, 17 Jan 2012 06:29:35 -0800 (PST) Received: from apple.bbn.com ([128.89.72.183]:49288) by smtp.bbn.com with esmtp (Exim 4.74 (FreeBSD)) (envelope-from ) id 1RnA2X-000AEK-FY; Tue, 17 Jan 2012 09:29:33 -0500 Mime-Version: 1.0 (Apple Message framework v1084) Content-Type: text/plain; charset=us-ascii From: John Zinky In-Reply-To: Date: Tue, 17 Jan 2012 09:29:33 -0500 Content-Transfer-Encoding: quoted-printable Message-Id: <0F578B93-EDAB-4722-A82E-8BDCD1A15C24@bbn.com> References: To: Razvan BEURAN X-Mailer: Apple Mail (2.1084) Cc: dtn-users@irtf.org Subject: Re: [dtn-users] dtnping performance issues X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 17 Jan 2012 14:29:36 -0000 Razan, It might be helpful to post the DTN 2.8 config files for both BPAs. That would answer all the config questions. (You might have to strip off the comments). thanks zinky On Jan 15, 2012, at 8:06 PM, Razvan BEURAN wrote: >=20 > Dear all, >=20 > I have started using DTN2 several months ago, and it has generally = been a positive experience. However, some of the experiments I did = produced some unexpected results in terms of performance, and I became = even more puzzled after doing similar experiments with another DTN = implementation most of you may already know, IBR-DTN. >=20 > Since it would be too lengthy to dicuss all the problems I had, I = shall focus here on a single and very simple issue, the RTT shown by = dtnping when run between two PCs: >=20 > * DTN2 dtnping (temporary storage in file, > persistent storage in BerkleyDB): 90 ms > * DTN2 dtnping (temporary storage in file, > persistent storage in memory): 41 ms > * IBR-DTN dtnping (temporary storage in file, > persistent storage in file): 4.2 ms > * IBR-DTN dtnping (temporary storage in file, > persistent storage in memory): 3.6 ms > * IBR-DTN dtnping (temporary storage in memory, > persistent storage in memory): 2.9 ms >=20 > FYI, the PCs are relatively old but fast enough (Dual-core Pentium 4, = 3.2 GHz, 4GB RAM) and are connected via a switch. The standard ping = command shows an RTT of about 1 ms between these 2 PCs. >=20 > The results for DTN2 get even worse when using a large number of nodes = (10 and 26), but since these large-experiment conditions are difficult = to reproduce I will skip this issue for the time being. >=20 > Given the above, I have two (sets of) questions: >=20 > 1) Are the DTN2 dtnping results I obtained normal? Can anyone provide > results from a similar experiment for comparison purposes. Has = anyone > done similar experiments with IBR-DTN? Are there any suggestions for > improving performance of DTN2? >=20 > 2) Has anyone done DTN2 experiments with a large number of nodes? By = large > I mean at least more than 10, but preferably more than 25. My > preliminary results with dtnping show quite a poor performance of = DTN2 > compared with IBR-DTN in large settings, and I can provide more = details > about this if requested. >=20 > Best wishes, > Razvan > _______________________________________________ > dtn-users mailing list > dtn-users@irtf.org > https://www.irtf.org/mailman/listinfo/dtn-users From oreilly@mbari.org Thu Jan 19 09:49:38 2012 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DEEC221F854C for ; Thu, 19 Jan 2012 09:49:38 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.714 X-Spam-Level: X-Spam-Status: No, score=-1.714 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_FONT_FACE_BAD=0.884, 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 nDrL4gu8dixx for ; Thu, 19 Jan 2012 09:49:38 -0800 (PST) Received: from gustnado.mbari.org (gustnado.mbari.org [134.89.2.12]) by ietfa.amsl.com (Postfix) with ESMTP id F1BFF21F86C4 for ; Thu, 19 Jan 2012 09:49:37 -0800 (PST) Received: from snow.shore.mbari.org (snow.shore.mbari.org [134.89.12.65]) by gustnado.mbari.org (8.13.8/8.13.8) with ESMTP id q0JHnboU016911 for ; Thu, 19 Jan 2012 09:49:37 -0800 (PST) Received: from slush.shore.mbari.org ([134.89.12.31]) by snow.shore.mbari.org with Microsoft SMTPSVC(6.0.3790.4675); Thu, 19 Jan 2012 09:49:37 -0800 Received: from localhost (localhost.localdomain [127.0.0.1]) by slush.shore.mbari.org (Postfix) with ESMTP id 47C9645E002 for ; Thu, 19 Jan 2012 09:49:37 -0800 (PST) X-Virus-Scanned: amavisd-new at mbari.org Received: from slush.shore.mbari.org ([127.0.0.1]) by localhost (slush.shore.mbari.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iGyNhcGNULAb for ; Thu, 19 Jan 2012 09:49:36 -0800 (PST) Received: from [134.89.8.141] (unknown [134.89.8.141]) by slush.shore.mbari.org (Postfix) with ESMTPSA id CBCFA45E001 for ; Thu, 19 Jan 2012 09:49:36 -0800 (PST) Message-Id: From: "Tom O'Reilly" To: dtn-users@irtf.org Content-Type: multipart/alternative; boundary=Apple-Mail-5-795825872 Mime-Version: 1.0 (Apple Message framework v936) Date: Thu, 19 Jan 2012 09:49:36 -0800 X-Mailer: Apple Mail (2.936) X-OriginalArrivalTime: 19 Jan 2012 17:49:37.0194 (UTC) FILETIME=[B601A8A0:01CCD6D2] Subject: [dtn-users] Broken documentation link X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 19 Jan 2012 17:49:39 -0000 --Apple-Mail-5-795825872 Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit The DTN2 documentation file manual/arch.html says: The following text is intended to provide a sketch of how the whole thing fits together. It is a brief synopsis of DTN2s implementation, taken entirely from this paper by Michael Demmer and others. For more information about DTN2's architecture and role as a platform for experimentation, see this paper. But that link referenced above (http://www.cs.berkeley.edu/~demmer/papers/dtn-irb-tr-04-020.pdf ) goes to a UC Berkley page that says "The server has encountered a problem because the resource was not found". After some searching, I presume that the Demming paper is at www.dtnrg.org/docs/papers/demmer-irb-tr-04-020.pdf - is this correct? Someone should fix the documentation link... Thanks, Tom --Apple-Mail-5-795825872 Content-Type: text/html; charset=US-ASCII Content-Transfer-Encoding: quoted-printable The DTN2 documentation file manual/arch.html = says:

The following text is intended = to provide a sketch of how the whole thing fits together. It is a brief synopsis of DTN2s=20 implementation, taken entirely from this paper by Michael Demmer and others. For more information about = DTN2's architecture and role as a platform for experimentation, see this paper.

But that link = referenced above (h= ttp://www.cs.berkeley.edu/~demmer/papers/dtn-irb-tr-04-020.pdf) goes = to a UC Berkley page that says "The server has encountered a problem = because the resource was not found".

After some searching, I presume that the Demming = paper is at www.dtnrg.org/docs/papers/demmer-irb-tr-04-020.pdf - is = this correct?

Someone should fix the = documentation link...

Thanks,
Tom

= --Apple-Mail-5-795825872-- From razvan@nict.go.jp Sun Jan 22 15:52:35 2012 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8732921F85B7 for ; Sun, 22 Jan 2012 15:52:35 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: 1.653 X-Spam-Level: * X-Spam-Status: No, score=1.653 tagged_above=-999 required=5 tests=[AWL=-0.857, BAYES_50=0.001, HELO_EQ_JP=1.244, HOST_EQ_JP=1.265] 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 PH1Xhet7+4n9 for ; Sun, 22 Jan 2012 15:52:34 -0800 (PST) Received: from mow.securemx.jp (mow302.securemx.jp [210.130.202.52]) by ietfa.amsl.com (Postfix) with ESMTP id ABB9721F85B5 for ; Sun, 22 Jan 2012 15:52:34 -0800 (PST) Received: by mow.securemx.jp (mow302) id q0MNqQPj004697; Mon, 23 Jan 2012 08:52:27 +0900 X-MXL-Hash: 4f1ca13a2214b05e-a4db0a41e285b28540cc3ba088aab1655c323cc2 Received: from ns2.nict.go.jp (ns2.nict.go.jp [133.243.3.2]) by relay.securemx.jp (mx-mr300) id q0MNqPcP011778 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Mon, 23 Jan 2012 08:52:25 +0900 Received: from gw2.nict.go.jp (gw2 [133.243.18.251]) by ns2.nict.go.jp with ESMTP id q0MNqPFs025843; Mon, 23 Jan 2012 08:52:25 +0900 (JST) Received: from gw2.nict.go.jp (localhost [127.0.0.1]) by gw2.nict.go.jp with ESMTP id q0MNqPXE029630; Mon, 23 Jan 2012 08:52:25 +0900 (JST) Received: from mail2.nict.go.jp (mail.nict.go.jp [133.243.18.3]) by gw2.nict.go.jp with ESMTP id q0MNqOcO029627; Mon, 23 Jan 2012 08:52:24 +0900 (JST) Received: from mail2.nict.go.jp (localhost [127.0.0.1]) by mail2.nict.go.jp (NICT Mail) with ESMTP id 9695416965; Mon, 23 Jan 2012 08:52:24 +0900 (JST) Received: from ssh (ssh.nict.go.jp [133.243.3.49]) by mail2.nict.go.jp (NICT Mail) with ESMTP id 90F8D16950; Mon, 23 Jan 2012 08:52:24 +0900 (JST) Date: Mon, 23 Jan 2012 08:52:24 +0900 (JST) From: Razvan BEURAN To: John Zinky In-Reply-To: <0F578B93-EDAB-4722-A82E-8BDCD1A15C24@bbn.com> Message-ID: References: <0F578B93-EDAB-4722-A82E-8BDCD1A15C24@bbn.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: dtn-users@irtf.org Subject: Re: [dtn-users] dtnping performance issues X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 22 Jan 2012 23:52:35 -0000 Dear all, Sorry for my late reply, but I was away from office for a while. In order to reduce speculations, I made an even simpler experiment that anyone should be able to reproduce. Thus, I used the default configuration file in the DTN2 distribution, and I did dtnping to localhost. Since the configuration file is the default one, I believe it is not necessary to include it. Here are the results I obtained for the latest DTN2 releases: 1) For dtn-2.8 > ./dtnping localhost source_eid [dtn://hpc1.starbed.net.dtn/ping.8026] dtn_register succeeded, regid 10 PING [dtn://hpc1.starbed.net.dtn/ping] (expiration 30)... 20 bytes from [dtn://hpc1.starbed.net.dtn/ping]: 'dtnping!' seqno=0, time=191 ms 20 bytes from [dtn://hpc1.starbed.net.dtn/ping]: 'dtnping!' seqno=1, time=168 ms 20 bytes from [dtn://hpc1.starbed.net.dtn/ping]: 'dtnping!' seqno=2, time=168 ms 20 bytes from [dtn://hpc1.starbed.net.dtn/ping]: 'dtnping!' seqno=3, time=201 ms 20 bytes from [dtn://hpc1.starbed.net.dtn/ping]: 'dtnping!' seqno=4, time=191 ms 20 bytes from [dtn://hpc1.starbed.net.dtn/ping]: 'dtnping!' seqno=5, time=176 ms 2) For dtn-2.7 > ./dtnping localhost source_eid [dtn://hpc1.starbed.net.dtn/ping.7986] dtn_register succeeded, regid 10 PING [dtn://hpc1.starbed.net.dtn/ping] (expiration 30)... 20 bytes from [dtn://hpc1.starbed.net.dtn/ping]: 'dtnping!' seqno=0, time=33 ms 20 bytes from [dtn://hpc1.starbed.net.dtn/ping]: 'dtnping!' seqno=1, time=35 ms 20 bytes from [dtn://hpc1.starbed.net.dtn/ping]: 'dtnping!' seqno=2, time=35 ms 20 bytes from [dtn://hpc1.starbed.net.dtn/ping]: 'dtnping!' seqno=3, time=36 ms 20 bytes from [dtn://hpc1.starbed.net.dtn/ping]: 'dtnping!' seqno=4, time=35 ms 20 bytes from [dtn://hpc1.starbed.net.dtn/ping]: 'dtnping!' seqno=5, time=35 ms 3) For comparison here are the ordinary ping results: > ping localhost PING localhost.localdomain (127.0.0.1) 56(84) bytes of data. 64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=1 ttl=64 time=0.033 ms 64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=2 ttl=64 time=0.027 ms 64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=3 ttl=64 time=0.027 ms 64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=4 ttl=64 time=0.027 ms 64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=5 ttl=64 time=0.027 ms My results show a 6x increase in delay for dtn-2.8 compared to dtn-2.7, and about 1000x increase in delay for dtn-2.7 compared to ordinary ping. These huge differences are very puzzling, and I would appreciate if someone could confirm whether the results I obtained are reproducible or not. If the results are reproducible, I believe this performance issue should be addressed somehow, otherwise large experiments cannot be performed. If the results are not reproducible, then it would be nice to be able to find the cause. By the way, I used for this experiments a high performance PC (Intel i7-2600 CPU @ 3.40GHz) and the OS was Scientific Linux 6.0 (a variant of RedHat Enterprise Linux 6.0). Best wishes, Razvan On Tue, 17 Jan 2012, John Zinky wrote: > Razan, > It might be helpful to post the DTN 2.8 config files for both BPAs. > That would answer all the config questions. > (You might have to strip off the comments). > > thanks > zinky > > On Jan 15, 2012, at 8:06 PM, Razvan BEURAN wrote: > >> >> Dear all, >> >> I have started using DTN2 several months ago, and it has generally been a positive experience. However, some of the experiments I did produced some unexpected results in terms of performance, and I became even more puzzled after doing similar experiments with another DTN implementation most of you may already know, IBR-DTN. >> >> Since it would be too lengthy to dicuss all the problems I had, I shall focus here on a single and very simple issue, the RTT shown by dtnping when run between two PCs: >> >> * DTN2 dtnping (temporary storage in file, >> persistent storage in BerkleyDB): 90 ms >> * DTN2 dtnping (temporary storage in file, >> persistent storage in memory): 41 ms >> * IBR-DTN dtnping (temporary storage in file, >> persistent storage in file): 4.2 ms >> * IBR-DTN dtnping (temporary storage in file, >> persistent storage in memory): 3.6 ms >> * IBR-DTN dtnping (temporary storage in memory, >> persistent storage in memory): 2.9 ms >> >> FYI, the PCs are relatively old but fast enough (Dual-core Pentium 4, 3.2 GHz, 4GB RAM) and are connected via a switch. The standard ping command shows an RTT of about 1 ms between these 2 PCs. >> >> The results for DTN2 get even worse when using a large number of nodes (10 and 26), but since these large-experiment conditions are difficult to reproduce I will skip this issue for the time being. >> >> Given the above, I have two (sets of) questions: >> >> 1) Are the DTN2 dtnping results I obtained normal? Can anyone provide >> results from a similar experiment for comparison purposes. Has anyone >> done similar experiments with IBR-DTN? Are there any suggestions for >> improving performance of DTN2? >> >> 2) Has anyone done DTN2 experiments with a large number of nodes? By large >> I mean at least more than 10, but preferably more than 25. My >> preliminary results with dtnping show quite a poor performance of DTN2 >> compared with IBR-DTN in large settings, and I can provide more details >> about this if requested. >> >> Best wishes, >> Razvan >> _______________________________________________ >> dtn-users mailing list >> dtn-users@irtf.org >> https://www.irtf.org/mailman/listinfo/dtn-users > >