From mboxrd@z Thu Jan 1 00:00:00 1970 From: Rick Jones Subject: Re: [RFC] [PATCH] Optimize TCP sendmsg in favour of fast devices? Date: Tue, 19 Jan 2010 09:41:26 -0800 Message-ID: <4B55EEC6.4070402@hp.com> References: <20100115053352.31564.765.sendpatchset@krkumar2.in.ibm.com> <4B50B032.2060609@hp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Cc: davem@davemloft.net, eric.dumazet@gmail.com, ilpo.jarvinen@helsinki.fi, netdev@vger.kernel.org To: Krishna Kumar2 Return-path: Received: from g1t0029.austin.hp.com ([15.216.28.36]:5179 "EHLO g1t0029.austin.hp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754575Ab0ASRl3 (ORCPT ); Tue, 19 Jan 2010 12:41:29 -0500 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: Krishna Kumar2 wrote: > Rick Jones wrote on 01/15/2010 11:43:06 PM: >>Multiple process tests may not be as easy in netperf as it is in >>iperf, but under: >> >>ftp://ftp.netperf.org/netperf/misc >> >> I have a single-stream test script I use called runemomni.sh and an >> example of its output, as well as an aggregate script I use called >> runemomniagg2.sh - I'll post an example of its output there as >> soon as I finish some runs. > > > I usually run netperf for smaller number of threads and aggregate > the output using some scripts. I will try what you suggested above, > and see if I can get consistent results for higher number of > processes. Thanks for the links. You're welcome - the output of the aggregate script is up there too now - generally I run three systems for the aggregate tests and then include the TCP_MAERTs stuff, but in this case since I had only two otherwise identical systems, TCP_MAERTs would have been redundant. rick jones