From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755890AbXJ2IaI (ORCPT ); Mon, 29 Oct 2007 04:30:08 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752084AbXJ2I34 (ORCPT ); Mon, 29 Oct 2007 04:29:56 -0400 Received: from web32611.mail.mud.yahoo.com ([68.142.207.238]:43209 "HELO web32611.mail.mud.yahoo.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1751581AbXJ2I3z (ORCPT ); Mon, 29 Oct 2007 04:29:55 -0400 X-YMail-OSG: Pbr4zfUVM1nHeJbM3glvxegEASGcsmP5b5ShRGzHlhRVSPVN X-Mailer: YahooMailRC/814.06 YahooMailWebService/0.7.152 Date: Mon, 29 Oct 2007 01:29:49 -0700 (PDT) From: Martin Knoblauch Subject: Re: 2.6.24-rc1: First impressions To: Andrew Morton , Arjan van de Ven Cc: Ingo Molnar , linux-kernel@vger.kernel.org, a.p.zijlstra@chello.nl, wfg@mail.ustc.edu.cn, torvalds@linux-foundation.org, riel@redhat.com MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Message-ID: <193426.68317.qm@web32611.mail.mud.yahoo.com> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org ----- Original Message ---- > From: Andrew Morton > To: Arjan van de Ven > Cc: Ingo Molnar ; spamtrap@knobisoft.de; linux-kernel@vger.kernel.org; a.p.zijlstra@chello.nl; wfg@mail.ustc.edu.cn; torvalds@linux-foundation.org; riel@redhat.com > Sent: Saturday, October 27, 2007 7:59:51 AM > Subject: Re: 2.6.24-rc1: First impressions > > On Fri, 26 Oct 2007 22:46:57 -0700 Arjan van de > Ven > wrote: > > > > > > dd1 - copy 16 GB from /dev/zero to local FS > > > > > dd1-dir - same, but using O_DIRECT for output > > > > > dd2/dd2-dir - copy 2x7.6 GB in parallel from /dev/zero to > local > FS > > > > > dd3/dd3-dir - copy 3x5.2 GB in parallel from /dev/zero lo > local > FS > > > > > net1 - copy 5.2 GB from NFS3 share to local FS > > > > > mix3 - copy 3x5.2 GB from /dev/zero to local disk and two NFS3 > > > > > shares > > > > > > > > > > I did the numbers for 2.6.19.2, 2.6.22.6 and 2.6.24-rc1. All > > > > > units are MB/sec. > > > > > > > > > > test 2.6.19.2 2.6.22.6 2.6.24.-rc1 > > > > > > > ---------------------------------------------------------------- > > > > > dd1 28 50 96 > > > > > dd1-dir 88 88 86 > > > > > dd2 2x16.5 2x11 2x44.5 > > > > > dd2-dir 2x44 2x44 2x43 > > > > > dd3 3x9.8 3x8.7 3x30 > > > > > dd3-dir 3x29.5 3x29.5 3x28.5 > > > > > net1 30-33 50-55 37-52 > > > > > mix3 17/32 25/50 96/35 > > > > > (disk/combined-network) > > > > > > > > wow, really nice results! > > > > > > Those changes seem suspiciously large to me. I wonder if > there's > less > > > physical IO happening during the timed run, and > correspondingly > more > > > afterwards. > > > > > > > another option... this is ext2.. didn't the ext2 reservation > stuff > get > > merged into -rc1? for ext3 that gave a 4x or so speed boost (much > > better sequential allocation pattern) > > > > Yes, one would expect that to make a large difference in > dd2/dd2-dir > and > dd3/dd3-dir - but only on SMP. On UP there's not enough concurrency > in the fs block allocator for any damage to occur. > Just for the record the test are done on SMP. > Reservations won't affect dd1 though, and that went faster too. > This is the one result that surprised me most, as I did not really expect any big moves here. I am not complaining :-), but definitely it would be nice to understand the why. Cheers Martin >