From owner-freebsd-net@FreeBSD.ORG Thu Mar 11 21:41:18 2004 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8A42316A4CE; Thu, 11 Mar 2004 21:41:18 -0800 (PST) Received: from rms04.rommon.net (rms04.rommon.net [212.54.2.140]) by mx1.FreeBSD.org (Postfix) with ESMTP id 70EAB43D2D; Thu, 11 Mar 2004 21:41:17 -0800 (PST) (envelope-from pete@he.iki.fi) Received: from he.iki.fi (h81.vuokselantie10.fi [193.64.42.129]) by rms04.rommon.net (8.12.9p1/8.12.9) with ESMTP id i2C5fEcM087064; Fri, 12 Mar 2004 07:41:14 +0200 (EET) (envelope-from pete@he.iki.fi) Message-ID: <40514D78.6020605@he.iki.fi> Date: Fri, 12 Mar 2004 07:41:12 +0200 From: Petri Helenius User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.6) Gecko/20040113 X-Accept-Language: en-us, en MIME-Version: 1.0 To: David Malone References: <20040310192255.GD14892@Odin.AC.HMC.Edu> <20040310193840.6479F77A6D4@guns.icir.org> <20040311225347.GA66644@walton.maths.tcd.ie> In-Reply-To: <20040311225347.GA66644@walton.maths.tcd.ie> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit cc: freebsd-net@freebsd.org cc: freebsd-current@freebsd.org cc: Mark Allman Subject: Re: Who wants SACK? (Re: was My planned work on networking stack) X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 12 Mar 2004 05:41:18 -0000 David Malone wrote: >Mind you, Petri originally asked about evidence for two machines >back-to-back, and 100ms is rather long for that (unless you're at >Steven Low's lab ;-) > > Another interesting figure which comes to mind is whether "bursty loss" is the usual way a multigigabit optical link loses IP packets or if the flipping of single bit hits only one packet. This influences the actual real life problem a lot and in my understanding of 8B/10B coding, itīs designed not to lose sync over a single bit error so with a probability of bit error every few minutes, hitting two in close succession (in the window) is unlikely. Some time ago we did experiments implementing FEC at IP layer to make the multimedia which run over the network zero loss. While doing the experiment we recognized that the clustered loss we saw was caused by software issues in routers, not at any transmission devices. Using somewhat deeper interleaving of packets solved the issue with this application. Pete