Re: [iccrg] Congestion Control in Data Centers and HPC/RDMA environments

Michael Welzl <michawe@ifi.uio.no> Mon, 21 October 2019 15:24 UTC

Return-Path: <michawe@ifi.uio.no>
X-Original-To: iccrg@ietfa.amsl.com
Delivered-To: iccrg@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A78141200E3 for <iccrg@ietfa.amsl.com>; Mon, 21 Oct 2019 08:24:19 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.198
X-Spam-Level:
X-Spam-Status: No, score=-4.198 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, MIME_QP_LONG_LINE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id yBGAPkLNWhkp for <iccrg@ietfa.amsl.com>; Mon, 21 Oct 2019 08:24:16 -0700 (PDT)
Received: from mail-out01.uio.no (mail-out01.uio.no [IPv6:2001:700:100:10::50]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3005E120020 for <iccrg@irtf.org>; Mon, 21 Oct 2019 08:24:16 -0700 (PDT)
Received: from mail-mx05.uio.no ([129.240.10.49]) by mail-out01.uio.no with esmtps (TLSv1.2:ECDHE-RSA-AES256-GCM-SHA384:256) (Exim 4.92.3) (envelope-from <michawe@ifi.uio.no>) id 1iMZXY-0000D2-VL; Mon, 21 Oct 2019 17:24:12 +0200
Received: from telia-b2e840-164.connect.netcom.no ([178.232.64.164] helo=[100.70.64.163]) by mail-mx05.uio.no with esmtpsa (TLSv1.2:ECDHE-RSA-AES256-GCM-SHA384:256) user michawe (Exim 4.92.3) (envelope-from <michawe@ifi.uio.no>) id 1iMZXX-0002tW-8P; Mon, 21 Oct 2019 17:24:12 +0200
Content-Type: multipart/alternative; boundary="Apple-Mail-C0BCBA29-5B1C-458A-A790-CB410B0D2400"
Content-Transfer-Encoding: 7bit
From: Michael Welzl <michawe@ifi.uio.no>
Mime-Version: 1.0 (1.0)
Date: Mon, 21 Oct 2019 17:24:10 +0200
Message-Id: <1A516C17-ACC4-4FAE-92E4-51DC344C6E91@ifi.uio.no>
References: <CAAMqZPs8P3wjFfbyomCOC=mU7o0PfcLOO1zDXr4E01HPtmCZcg@mail.gmail.com>
Cc: iccrg IRTF list <iccrg@irtf.org>
In-Reply-To: <CAAMqZPs8P3wjFfbyomCOC=mU7o0PfcLOO1zDXr4E01HPtmCZcg@mail.gmail.com>
To: Paul Congdon <paul.congdon@tallac.com>
X-Mailer: iPhone Mail (17A878)
X-UiO-SPF-Received: Received-SPF: neutral (mail-mx05.uio.no: 178.232.64.164 is neither permitted nor denied by domain of ifi.uio.no) client-ip=178.232.64.164; envelope-from=michawe@ifi.uio.no; helo=[100.70.64.163];
X-UiO-Spam-info: not spam, SpamAssassin (score=-5.0, required=5.0, autolearn=disabled, AWL=0.045, HTML_MESSAGE=0.001, MIME_QP_LONG_LINE=0.001, UIO_MAIL_IS_INTERNAL=-5, uiobl=NO, uiouri=NO)
X-UiO-Scanned: 2CBA32509777DD4048D9CD6EFA54341F7E7F0023
Archived-At: <https://mailarchive.ietf.org/arch/msg/iccrg/CB2DiejanDMqwExM5gzARVhp-JU>
Subject: Re: [iccrg] Congestion Control in Data Centers and HPC/RDMA environments
X-BeenThere: iccrg@irtf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: "Discussions of Internet Congestion Control Research Group \(ICCRG\)" <iccrg.irtf.org>
List-Unsubscribe: <https://www.irtf.org/mailman/options/iccrg>, <mailto:iccrg-request@irtf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/iccrg/>
List-Post: <mailto:iccrg@irtf.org>
List-Help: <mailto:iccrg-request@irtf.org?subject=help>
List-Subscribe: <https://www.irtf.org/mailman/listinfo/iccrg>, <mailto:iccrg-request@irtf.org?subject=subscribe>
X-List-Received-Date: Mon, 21 Oct 2019 15:24:20 -0000

Yes, I’d say so. That was the point of adding this text when we did (many years ago).

Cheers
Michael 

Sent from my iPhone

> On 21 Oct 2019, at 17:11, Paul Congdon <paul.congdon@tallac.com> wrote:
> 
> 
> Yes, you are correct.
> 
> "Because congestion control is typically a key function of transport protocols, it is hard to separate other transport layer issues from congestion control. Considerations for deployment of transport protocols are therefore also considered within scope. The ICCRG may also consider congestion and protocol performance problems in general IP networks, i.e., not only on the global Internet. One example of such IP networks are multi-tenant, heterogeneous datacenters, which in some sense represent a microcosm of the larger Internet; changes to the transport protocols being considered in that space are fundamental and can have broader implications."
> 
> One question that came up in discussions with others is can the solution be restricted to particular topologies and domains, such as datacenters, if the solution is not deemed 'safe' to run across the Internet at large?
> 
> Sounds like such research is in scope from the above paragraph.
> 
> 
>> On Thu, Oct 17, 2019 at 10:31 PM Michael Welzl <michawe@ifi.uio.no> wrote:
>> Hi,
>> 
>> What about adding a statement like the following to the charter?
>> 
>> ***
>> The ICCRG may also consider congestion and protocol performance problems in general IP networks, i.e., not only on the global Internet. One example of such IP networks are multi-tenant, heterogeneous datacenters, which in some sense represent a microcosm of the larger Internet; changes to the transport protocols being considered in that space are fundamental and can have broader implications.
>> ***
>> 
>> (turns out that it’s already there:  https://datatracker.ietf.org/rg/iccrg/about/ )
>> 
>> Cheers,
>> Michael
>> 
>> 
>>> On Oct 17, 2019, at 7:54 PM, Paul Congdon <paul.congdon@tallac.com> wrote:
>>> 
>>> Hello ICCRG,
>>> 
>>> Jana suggested that I email the group with this call for interest on Congestion Control for Data Centers.  At the last several IETF meetings I've hosted side meetings and presented to various groups about work going on in Data Center congestion control and scaling HPC/RDMA networks.  We have been searching for a permanent venue within the IETF for this work.   ICCRG is primarily focused on end-to-end congestion control across the vast and expansive Internet.   The data center, and in particular, the HPC/RDMA data center network has much different characteristics.   So, the question to ICCRG is this; is this topic of interest to ICCRG and should the charter be expanded to include DC congestion control, or is this a good topic for a new research item in IRTF?
>>> 
>>> Looking forward to hearing from you,
>>> Thank you,
>>> Paul Congdon
>>> 
>>> _______________________________________________
>>> iccrg mailing list
>>> iccrg@irtf.org
>>> https://www.irtf.org/mailman/listinfo/iccrg
>> 
>> _______________________________________________
>> iccrg mailing list
>> iccrg@irtf.org
>> https://www.irtf.org/mailman/listinfo/iccrg