From: Alexey Kuznetsov Subject: Re: [PATCH][RFC] Re: high latency with TCP connections Date: Tue, 19 Sep 2006 00:41:05 +0400 Message-ID: <20060918204105.GA31333@ms2.inr.ac.ru> References: <20060830100734.GA22235@isil.ipib.msu.ru> <20060904160045.GA15599@ms2.inr.ac.ru> <44FDBA04.3080104@hp.com> <20060918.003938.26278728.davem@davemloft.net> <450ED337.8000700@hp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: David Miller , alex@sectorb.msk.ru, netdev@vger.kernel.org Return-path: Received: from minus.inr.ac.ru ([194.67.69.97]:16294 "HELO ms2.inr.ac.ru") by vger.kernel.org with SMTP id S964959AbWIRUlt (ORCPT ); Mon, 18 Sep 2006 16:41:49 -0400 To: Rick Jones Content-Disposition: inline In-Reply-To: <450ED337.8000700@hp.com> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Hello! Of course, number of ACK increases. It is the goal. :-) > unpleasant increase in service demands on something like a "burst > enabled" (./configure --enable-burst) netperf TCP_RR test: > > netperf -t TCP_RR -H foo -- -b N # N > 1 foo=localhost b patched orig 2 105874.83 105143.71 3 114208.53 114023.07 4 120493.99 120851.27 5 128087.48 128573.33 10 151328.48 151056.00 Probably, the test is done wrong. But I see no difference. > to increase as a result. Pipelined HTTP would be like that, some NFS > over TCP stuff too, maybe X traffic, X will be excited about better latency. What's about protocols not interested in latency, they will be a little happier, if transactions are processed asynchronously. But actually, it is not about increasing/decreasing number of ACKs. It is about killing that pain in ass which we used to have because we pretended to be too smart. Alexey