From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: TCP rx window autotuning harmful at LAN context Date: Tue, 10 Mar 2009 06:19:27 +0100 Message-ID: <49B5F85F.1080508@cosmosbay.com> References: <1e41a3230903091101u536a3b3bv7f0dd9da6891781e@mail.gmail.com> <20090309195906.M50328@bts.sk> <1e41a3230903091323j541d1895j2eb69b9f9c11f2f3@mail.gmail.com> <20090309.165205.11054081.davem@davemloft.net> <1e41a3230903091709i740d95ddhefc1df2c71ea60b6@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: David Miller , md@bts.sk, netdev@vger.kernel.org To: John Heffner Return-path: Received: from gw1.cosmosbay.com ([212.99.114.194]:53061 "EHLO gw1.cosmosbay.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752530AbZCJFTg convert rfc822-to-8bit (ORCPT ); Tue, 10 Mar 2009 01:19:36 -0400 In-Reply-To: <1e41a3230903091709i740d95ddhefc1df2c71ea60b6@mail.gmail.com> Sender: netdev-owner@vger.kernel.org List-ID: John Heffner a =E9crit : > On Mon, Mar 9, 2009 at 4:52 PM, David Miller wr= ote: >> From: John Heffner >> Date: Mon, 9 Mar 2009 13:23:15 -0700 >> >>> A drop-tail queue size of 1000 packets on a local interface is >>> questionable, and I think this is the real source of your problem. >> Are you suggested we decrease it? :-) >=20 > I am not so bold. :-D (And note the drop-tail prefix.) >=20 > A long queue with AQM would probably be best, but would require > careful testing before enabling by default. It would almost certainl= y > cause pain for some. >=20 > And, for the vast majority of people for whom the local interface is > not the bottleneck, it makes no difference. It hurts worst for > someone doing bulk transfer with a GigE device in 100 Mbps (or worse, > 10-Mbps) mode, where 1000 pkts is a long time, while simultaneously > doing something latency-sensitive. I suspect this is the case Marian > is experiencing. >=20 Interesting stuff indeed. Could you tell us more about AQM ?