From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vivek Goyal Subject: Re: [RFC] IO scheduler based IO controller V6 Date: Mon, 27 Jul 2009 23:27:12 -0400 Message-ID: <20090728032712.GC3620@redhat.com> References: <1246564917-19603-1-git-send-email-vgoyal@redhat.com> <4A6D0C9A.3080600@cn.fujitsu.com> <20090727125503.GA24449@redhat.com> Reply-To: device-mapper development Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Return-path: Content-Disposition: inline In-Reply-To: <20090727125503.GA24449@redhat.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: dm-devel-bounces@redhat.com Errors-To: dm-devel-bounces@redhat.com To: Gui Jianfeng Cc: dhaval@linux.vnet.ibm.com, snitzer@redhat.com, peterz@infradead.org, dm-devel@redhat.com, dpshah@google.com, jens.axboe@oracle.com, agk@redhat.com, balbir@linux.vnet.ibm.com, paolo.valente@unimore.it, fernando@oss.ntt.co.jp, mikew@google.com, jmoyer@redhat.com, nauman@google.com, m-ikeda@ds.jp.nec.com, lizf@cn.fujitsu.com, fchecconi@gmail.com, akpm@linux-foundation.org, jbaron@redhat.com, linux-kernel@vger.kernel.org, s-uchida@ap.jp.nec.com, righi.andrea@gmail.com, containers@lists.linux-foundation.org List-Id: dm-devel.ids On Mon, Jul 27, 2009 at 08:55:03AM -0400, Vivek Goyal wrote: > On Mon, Jul 27, 2009 at 10:10:34AM +0800, Gui Jianfeng wrote: > > Hi=EF=BC=8C > >=20 > > Here are some fio test results for IO Controller V6 built and not bui= lt. > > Iozone test results are also attached. > >=20 >=20 > Hi Gui, >=20 > Thanks a lot for some performance numbers. It seems to be a mixed chart= . > Performance gains at some places and loss at others. I am curious about > that -7.0% for normal writes. Not sure what can contribute to that. >=20 Gui, can you also try with CONFIG_TRACK_ASYNC_CONTEXT=3Dn and see if it i= mproves buffered write performance. Thanks Vivek > What was the value of "fairness" parameter when you ran those tests? Ca= n you > please set fairness =3D 0 and re-run the tests (if you have already not= done so). >=20 > By default fairness is set to 1 in V6. With fairness =3D 0, we should b= e very > close to existing CFQ behavior. If not, then we need to dive deeper and > see why variations are happening. >=20 > Is it also possible to run the same tests with V7.=20 >=20 > Thanks > Vivek >=20 > > Arch: X86 > > Mem: 1G > > Disk: 320G > > IO Scheduler: CFQ > >=20 > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > > By normal read and write syscall. > >=20 > > Block Size:32K > > File Size: 1G * 10 > >=20 > > Mode Normal read | Random read | Normal wr= ite | Random write | Direct read | Direct write > >=20 > > 2.6.31-rc1 47,932KiB/s 3,566KiB/s 45,693KiB= /s 8,501KiB/s 50,088KiB/s 43,473KiB/s > >=20 > > 2.6.31-rc1-Vivek-V6 47,231KiB/s 3,411KiB/s 42,451KiB= /s 8,714KiB/s 51,284KiB/s 42,341KiB/s > >=20 > > Performance -1.5% -4.4% -7.0% = +2.5% +2.4% -2.6% > >=20 > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > >=20 > > By mmap. > >=20 > > Block Size:32K > > File Size: 500M > >=20 > > Mode Normal read | Random read | Normal wr= ite | Random write > >=20 > > 2.6.31-rc1 49,951KiB/s 3,245KiB/s 21,950KiB= /s 2,771KiB/s > >=20 > > 2.6.31-rc1-Vivek-V6 49,951KiB/s 3,154KiB/s 22,593KiB= /s 2,648KiB/s > >=20 > > Performance 0% -2.8% +2.9% = -4.4% > >=20 > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > > =20 > > By libaio calls > >=20 > > Block Size:32K > > File Size: 500M > >=20 > > Mode Normal read | Random read | Normal wri= te | Random write > >=20 > > 2.6.31-rc1 49,447KiB/s 3,296KiB/s 57,519KiB= /s 21,093KiB/s > >=20 > > 2.6.31-rc1-Vivek-V6 50,142KiB/s 3,238KiB/s 57,791KiB= /s 21,283KiB/s > >=20 > > Performance +1.4% -1.8% 0% = +0.1% > >=20 > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > >=20 > >=20 > >=20 > >=20 > >=20 > >=20 > >=20 > >=20 > >=20 > >=20 > >=20 >=20 >=20 >=20 From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753399AbZG1D3N (ORCPT ); Mon, 27 Jul 2009 23:29:13 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753348AbZG1D3N (ORCPT ); Mon, 27 Jul 2009 23:29:13 -0400 Received: from mx2.redhat.com ([66.187.237.31]:39763 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752693AbZG1D3M (ORCPT ); Mon, 27 Jul 2009 23:29:12 -0400 Date: Mon, 27 Jul 2009 23:27:12 -0400 From: Vivek Goyal To: Gui Jianfeng Cc: linux-kernel@vger.kernel.org, containers@lists.linux-foundation.org, dm-devel@redhat.com, jens.axboe@oracle.com, nauman@google.com, dpshah@google.com, lizf@cn.fujitsu.com, mikew@google.com, fchecconi@gmail.com, paolo.valente@unimore.it, ryov@valinux.co.jp, fernando@oss.ntt.co.jp, s-uchida@ap.jp.nec.com, taka@valinux.co.jp, jmoyer@redhat.com, dhaval@linux.vnet.ibm.com, balbir@linux.vnet.ibm.com, righi.andrea@gmail.com, m-ikeda@ds.jp.nec.com, jbaron@redhat.com, agk@redhat.com, snitzer@redhat.com, akpm@linux-foundation.org, peterz@infradead.org Subject: Re: [RFC] IO scheduler based IO controller V6 Message-ID: <20090728032712.GC3620@redhat.com> References: <1246564917-19603-1-git-send-email-vgoyal@redhat.com> <4A6D0C9A.3080600@cn.fujitsu.com> <20090727125503.GA24449@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20090727125503.GA24449@redhat.com> User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Jul 27, 2009 at 08:55:03AM -0400, Vivek Goyal wrote: > On Mon, Jul 27, 2009 at 10:10:34AM +0800, Gui Jianfeng wrote: > > Hi， > > > > Here are some fio test results for IO Controller V6 built and not built. > > Iozone test results are also attached. > > > > Hi Gui, > > Thanks a lot for some performance numbers. It seems to be a mixed chart. > Performance gains at some places and loss at others. I am curious about > that -7.0% for normal writes. Not sure what can contribute to that. > Gui, can you also try with CONFIG_TRACK_ASYNC_CONTEXT=n and see if it improves buffered write performance. Thanks Vivek > What was the value of "fairness" parameter when you ran those tests? Can you > please set fairness = 0 and re-run the tests (if you have already not done so). > > By default fairness is set to 1 in V6. With fairness = 0, we should be very > close to existing CFQ behavior. If not, then we need to dive deeper and > see why variations are happening. > > Is it also possible to run the same tests with V7. > > Thanks > Vivek > > > Arch: X86 > > Mem: 1G > > Disk: 320G > > IO Scheduler: CFQ > > > > ============ > > By normal read and write syscall. > > > > Block Size:32K > > File Size: 1G * 10 > > > > Mode Normal read | Random read | Normal write | Random write | Direct read | Direct write > > > > 2.6.31-rc1 47,932KiB/s 3,566KiB/s 45,693KiB/s 8,501KiB/s 50,088KiB/s 43,473KiB/s > > > > 2.6.31-rc1-Vivek-V6 47,231KiB/s 3,411KiB/s 42,451KiB/s 8,714KiB/s 51,284KiB/s 42,341KiB/s > > > > Performance -1.5% -4.4% -7.0% +2.5% +2.4% -2.6% > > > > ============ > > > > By mmap. > > > > Block Size:32K > > File Size: 500M > > > > Mode Normal read | Random read | Normal write | Random write > > > > 2.6.31-rc1 49,951KiB/s 3,245KiB/s 21,950KiB/s 2,771KiB/s > > > > 2.6.31-rc1-Vivek-V6 49,951KiB/s 3,154KiB/s 22,593KiB/s 2,648KiB/s > > > > Performance 0% -2.8% +2.9% -4.4% > > > > ============ > > > > By libaio calls > > > > Block Size:32K > > File Size: 500M > > > > Mode Normal read | Random read | Normal write | Random write > > > > 2.6.31-rc1 49,447KiB/s 3,296KiB/s 57,519KiB/s 21,093KiB/s > > > > 2.6.31-rc1-Vivek-V6 50,142KiB/s 3,238KiB/s 57,791KiB/s 21,283KiB/s > > > > Performance +1.4% -1.8% 0% +0.1% > > > > ============ > > > > > > > > > > > > > > > > > > > > > > > > >