Re: [ledbat] list of reasons for needing multiple TCP connections

Joe Touch <touch@ISI.EDU> Fri, 21 November 2008 00:28 UTC

Return-Path: <ledbat-bounces@ietf.org>
X-Original-To: tana-archive@ietf.org
Delivered-To: ietfarch-tana-archive@core3.amsl.com
Received: from [127.0.0.1] (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 610A73A6801; Thu, 20 Nov 2008 16:28:33 -0800 (PST)
X-Original-To: ledbat@core3.amsl.com
Delivered-To: ledbat@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 2632B3A6801 for <ledbat@core3.amsl.com>; Thu, 20 Nov 2008 16:28:32 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.28
X-Spam-Level:
X-Spam-Status: No, score=-2.28 tagged_above=-999 required=5 tests=[AWL=-0.281, BAYES_00=-2.599, J_CHICKENPOX_25=0.6]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0tNuXIQ2-q22 for <ledbat@core3.amsl.com>; Thu, 20 Nov 2008 16:28:31 -0800 (PST)
Received: from vapor.isi.edu (vapor.isi.edu [128.9.64.64]) by core3.amsl.com (Postfix) with ESMTP id 2D5653A67F0 for <ledbat@ietf.org>; Thu, 20 Nov 2008 16:28:31 -0800 (PST)
Received: from [130.129.94.243] ([130.129.94.243]) by vapor.isi.edu (8.13.8/8.13.8) with ESMTP id mAL0SEet014991 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Thu, 20 Nov 2008 16:28:17 -0800 (PST)
Message-ID: <4926009E.2090103@isi.edu>
Date: Thu, 20 Nov 2008 16:28:14 -0800
From: Joe Touch <touch@ISI.EDU>
User-Agent: Thunderbird 2.0.0.18 (Windows/20081105)
MIME-Version: 1.0
To: Robb Topolski <robb@funchords.com>
References: <A0988C2F192F124FAFE3D70264C04587077A130C@xmb-sjc-231.amer.cisco.com> <C54B2BA4.3B85%lars.eggert@nokia.com> <3efc39a60811201424l5fc9f1d1keab7e811b5a8bca8@mail.gmail.com> <4925E805.1050801@isi.edu> <4925EAA5.9090107@asomi.com> <3efc39a60811201523w5cc95b21ya43b1224477c9c1d@mail.gmail.com>
In-Reply-To: <3efc39a60811201523w5cc95b21ya43b1224477c9c1d@mail.gmail.com>
X-Enigmail-Version: 0.95.7
X-ISI-4-43-8-MailScanner: Found to be clean
X-MailScanner-From: touch@isi.edu
Cc: ledbat@ietf.org
Subject: Re: [ledbat] list of reasons for needing multiple TCP connections
X-BeenThere: ledbat@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Mailing list of the LEDBAT WG <ledbat.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/ledbat>, <mailto:ledbat-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/pipermail/ledbat>
List-Post: <mailto:ledbat@ietf.org>
List-Help: <mailto:ledbat-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/ledbat>, <mailto:ledbat-request@ietf.org?subject=subscribe>
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Sender: ledbat-bounces@ietf.org
Errors-To: ledbat-bounces@ietf.org

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi, all,

It's useful to differentiate between TCP not getting full BW because it
uses only a single path, vs. TCP not getting full BW because its window
doesn't open fast enough.

SCTP addresses the former, not the latter; the two have the same problem
in the latter case. There are a variety of extensions to TCP's windowing
mechanism to address high BW*delay product environments which apply
equally well to SCTP in those cases.

Joe

Robb Topolski wrote:
> 
> 
> On Thu, Nov 20, 2008 at 2:54 PM, Caitlin Bestler <cait@asomi.com
> <mailto:cait@asomi.com>> wrote:
> 
>     As a point of reference, part of the justification for doing iSCSI
>     over multiple TCP connections within a session was because "It
>     might be impossible to achieve the full bandwidth capability
>     of the underlying physical transport by using a single TCP
>     connection."
> 
>     [http://storageconference.org/2003/papers/19-Meth-Design.pdf
>     section 2.2]
> 
> 
> Good stuff there, the relevant reasoning that might be sucked into a
> LEDBAT paper include:
> 
>               It might be impossible to achieve the full bandwidth capa-
>               bility of the underlying physical transport by using a sin-
>               gle TCP connection (possibly due to the TCP window
>               size and the round-trip-time of TCP acknowledgements
>               over long distances). Some protocols, like SCTP, auto-
>               matically and transparently distribute their traffic over a
>               number of connections in order to achieve the full band-
>               width. There may also be several physical interconnects
>               (i.e. separate cables) connecting the initiator and target,
>               and it would be most desirable to aggregate and simulta-
>               neously utilize all such existing physical connections.
>  
> 
> 
> -- 
> Robb Topolski (robb@funchords.com <mailto:robb@funchords.com>)
> Hillsboro, Oregon USA
> http://www.funchords.com/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkkmAJ4ACgkQE5f5cImnZrs6VACfZ/dvo9SUC8UWzKZHJTx0NGah
ATQAoIbqSiD8KCzZ1W3JDg10v5qevxqL
=AHaI
-----END PGP SIGNATURE-----
_______________________________________________
ledbat mailing list
ledbat@ietf.org
https://www.ietf.org/mailman/listinfo/ledbat