From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-13-ewr.dyndns.com (mxout-137-ewr.mailhop.org [216.146.33.137]) by lists.bufferbloat.net (Postfix) with ESMTP id 73A542E001D for ; Mon, 4 Apr 2011 06:20:14 -0700 (PDT) Received: from scan-11-ewr.mailhop.org (scan-11-ewr.local [10.0.141.229]) by mail-13-ewr.dyndns.com (Postfix) with ESMTP id BE0A0A4A439 for ; Mon, 4 Apr 2011 13:20:13 +0000 (UTC) X-Spam-Score: 0.0 () X-Mail-Handler: MailHop by DynDNS X-Originating-IP: 149.20.54.64 Received: from mainmail.teklibre.com (toutatis.isc.org [149.20.54.64]) by mail-13-ewr.dyndns.com (Postfix) with ESMTP id 00093A4F594 for ; Mon, 4 Apr 2011 13:20:05 +0000 (UTC) Received: from localhost (localhost [127.0.0.1]) by mainmail.teklibre.com (Postfix) with ESMTP id 0AEB412B75B for ; Mon, 4 Apr 2011 06:53:49 -0700 (PDT) Received: from mainmail.teklibre.com ([127.0.0.1]) by localhost (toutatis.sql1.isc.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UwUyfEcit1Gn for ; Mon, 4 Apr 2011 06:53:48 -0700 (PDT) Received: from [143.215.109.199] (lawn-143-215-109-199.lawn.gatech.edu [143.215.109.199]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) (Authenticated sender: d) by mainmail.teklibre.com (Postfix) with ESMTPSA id 881BD12B756 for ; Mon, 4 Apr 2011 06:53:48 -0700 (PDT) Message-ID: <4D99C57F.9070105@teklibre.com> Date: Mon, 04 Apr 2011 07:19:59 -0600 From: Dave Taht User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.14) Gecko/20110223 Thunderbird/3.1.8 MIME-Version: 1.0 To: bismark-devel@lists.bufferbloat.net Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit Subject: [Bismark-devel] Alternate test tools X-BeenThere: bismark-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 04 Apr 2011 13:20:14 -0000 Sri tells me that iperf has issues with socket reuse. Perhaps nttcp or netperf would be better? http://linux.die.net/man/1/nttcp OR svn co http://www.netperf.org/svn/netperf2/trunk/ These were the experiments I was running with rick jones (of hp) on tcp window sizing using netperf: Dave Täht wrote: >>Sadly, I see Linux's dynamic window sizing take the window to 4MB when >>128KB would do. I'm not familiar with the behaviour of the other >>stacks' > > How can I replicate your results? Grab top-of-trunk netperf with svn from: http://www.netperf.org/svn/netperf2/trunk cd to where you put the bits and: ./configure --enable-omni --enable-burst --enable-histogram ... make need to do the same on your other system, and launch netserver there. Then back on the first system something like: netperf -t TCP_STREAM -H -- -O LSS_SIZE,LSS_SIZE_END,RSR_SIZE,RSR_SIZE_END and you should see output similar to: raj@tardy:~/netperf2_trunk$ src/netperf -t TCP_STREAM -H localhost -- -O LSS_SIZE,LSS_SIZE_END,RSR_SIZE,RSR_SIZE_E> MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to localhost (127.0.0.1) port 0 AF_INET : histogram Local Local Remote Remote Send Socket Send Socket Recv Socket Recv Socket Size Size Size Size Initial Final Initial Final 16384 237272 87380 4194304 Replace the "test specific" (in the section after the "--") -O with -o it will be CSV output. If you replace it with -k it will be keyval output. If you give it a ? as the value it will emit all the choices. You can put a list of desired output into a file and pass that as the option to -O, -o or -k Here is a run between two systems on in Palo Alto, CA and the other in Cupertino raj@tardy:~/netperf2_trunk$ src/netperf -t TCP_STREAM -H 16.89.247.36 -- -k THROUGHPUT,LSS_SIZE,LSS_SIZE_END,RSR_S> MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 16.89.247.36 (16.89.247.36) port 0 AF_INET : histogram THROUGHPUT=189.75 LSS_SIZE=16384 LSS_SIZE_END=4194304 RSR_SIZE=87380 RSR_SIZE_END=4194304 Normally the RTT between the two is O(7ms): 10 packets transmitted, 10 received, 0% packet loss, time 9039ms rtt min/avg/max/mdev = 6.961/7.053/7.249/0.122 ms