[Semi-OT] Transport resistance to reordering

"Spencer Dawkins" <spencer@mcsr-labs.org> Wed, 05 October 2005 17:43 UTC

Envelope-to: shim6-data@psg.com
Delivery-date: Wed, 05 Oct 2005 17:44:27 +0000
Message-ID: <177001c5c9d4$54676880$0500a8c0@china.huawei.com>
From: Spencer Dawkins <spencer@mcsr-labs.org>
To: shim6-wg <shim6@psg.com>
Cc: Mark Allman <mallman@icir.org>
Subject: [Semi-OT] Transport resistance to reordering
Date: Wed, 05 Oct 2005 12:43:33 -0500
MIME-Version: 1.0
Content-Type: text/plain; format="flowed"; charset="iso-8859-1"; reply-type="original"
Content-Transfer-Encoding: 7bit

We've had a few conversations about load-spreading that assumed TCP (and 
related transports) would respond badly to significant levels of reordering 
(they respond badly today).

I was chatting about this with Mark Allman, who pointed me towards 
http://www.ietf.org/internet-drafts/draft-ietf-tcpm-tcp-dcr-05.txt.

I don't think this is ON topic for Shim6, but if people are interested, it's 
on-topic for tcpm (http://www.ietf.org/html.charters/tcpm-charter.html).

(Since IP stacks have to be updated for Shim6 anyway...)

Thanks,

Spencer

[Semi-OT] Transport resistance to reordering Spencer Dawkins