From mboxrd@z Thu Jan 1 00:00:00 1970 From: John Heffner Subject: Re: TCP rx window autotuning harmful at LAN context Date: Tue, 10 Mar 2009 08:23:34 -0700 Message-ID: <1e41a3230903100823h5d5768d9wb9f36eb294840814@mail.gmail.com> References: <20090309195906.M50328@bts.sk> <1e41a3230903091323j541d1895j2eb69b9f9c11f2f3@mail.gmail.com> <20090310104956.GA81181@bts.sk> <20090310.043019.132650585.davem@davemloft.net> <20090310114606.GA84964@bts.sk> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-2 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: David Miller , netdev@vger.kernel.org To: =?ISO-8859-2?Q?Marian_=CFurkovi=E8?= Return-path: Received: from yx-out-2324.google.com ([74.125.44.30]:58689 "EHLO yx-out-2324.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756599AbZCJPXh convert rfc822-to-8bit (ORCPT ); Tue, 10 Mar 2009 11:23:37 -0400 Received: by yx-out-2324.google.com with SMTP id 8so1493722yxm.1 for ; Tue, 10 Mar 2009 08:23:35 -0700 (PDT) In-Reply-To: <20090310114606.GA84964@bts.sk> Sender: netdev-owner@vger.kernel.org List-ID: On Tue, Mar 10, 2009 at 4:46 AM, Marian =CFurkovi=E8 wrote: > On Tue, Mar 10, 2009 at 04:30:19AM -0700, David Miller wrote: >> From: Marian =CFurkovi=E8 >> Date: Tue, 10 Mar 2009 11:49:56 +0100 >> >> > Sender does not have the relevant info to implement this - it migh= t be >> > connected by 10 GE to the highspeed backbone. >> >> Yes, the sender does indeed have this information, and using it is >> exactly what congestion control algorithms such as VEGAS try to do. >> >> They look at both round trip times and bandwith as they increase the >> send congestion window. =A0And if round trips increase without a >> corresponding increase in bandwidth, they stop increasing. > > Yes, but that's actual bandwidth between sender and receiver, not > the hard BW limit of the receiver's NIC. My intention is just to intr= oduce > some safety belt preventing autotuning to increase the rx window > into MB ranges when RTT is very low. Nowhere in our proposal do you use NIC bandwidth. What you proposed can be done easily at the sender. -John