Date: Fri, 13 Jul 2001 11:02:28 -0400 From: Leo Bicknell <bicknell@ufp.org> To: Dan Nelson <dnelson@emsphone.com> Cc: Leo Bicknell <bicknell@ufp.org>, Paul Robinson <paul@akita.co.uk>, Julian Elischer <julian@elischer.org>, freebsd-hackers@FreeBSD.ORG Subject: Re: Network performance tuning. Message-ID: <20010713110228.A13674@ussenterprise.ufp.org> In-Reply-To: <20010713095228.C222@dan.emsphone.com>; from dnelson@emsphone.com on Fri, Jul 13, 2001 at 09:52:28AM -0500 References: <20010711195021.A89324@ussenterprise.ufp.org> <Pine.BSF.4.21.0107111904220.60496-100000@InterJet.elischer.org> <20010712175539.B93119@jake.akitanet.co.uk> <20010712210944.A73446@ussenterprise.ufp.org> <20010713095228.C222@dan.emsphone.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Jul 13, 2001 at 09:52:28AM -0500, Dan Nelson wrote: > Considering that w2k and Linux both have sack enabled by default, it's > not going away. Do you have a link to the thread that says sack > doesn't help? The best I can find is at the bottom of http://ftp.ee.lbl.gov/floyd/sacks.html which has http://www.cs.Berkeley.edu/~hari/papers/csd-97-966.ps on it. They do not seem to suggest that SACK _decreases_ performance, but they suggest from real world data capture that it would only help on 4% of the connections in that experience congestion in a production enviornment, and is thus probably not the best solution. My take it is that it's the best solution we have today, and best means "it helps in very few cases". One could argue it would be better to find a solution that works in more cases, but for now I would think SACK should be implemented so we have the best known solution. -- Leo Bicknell - bicknell@ufp.org Systems Engineer - Internetworking Engineer - CCIE 3440 Read TMBG List - tmbg-list-request@tmbg.org, www.tmbg.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010713110228.A13674>