From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:40217) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Z5yAO-000294-Sj for qemu-devel@nongnu.org; Fri, 19 Jun 2015 11:25:17 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Z5yAJ-0000g7-Tb for qemu-devel@nongnu.org; Fri, 19 Jun 2015 11:25:16 -0400 Received: from mail-wg0-x22c.google.com ([2a00:1450:400c:c00::22c]:36599) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Z5yAG-0000Yd-BG for qemu-devel@nongnu.org; Fri, 19 Jun 2015 11:25:11 -0400 Received: by wguu7 with SMTP id u7so20326447wgu.3 for ; Fri, 19 Jun 2015 08:25:01 -0700 (PDT) Date: Fri, 19 Jun 2015 16:24:58 +0100 From: Stefan Hajnoczi Message-ID: <20150619152458.GC20966@stefanha-thinkpad.redhat.com> References: <7c4c87b8.111fa.14df75bb798.Coremail.cauchy-love@163.com> <20150616130539.GC29405@stefanha-thinkpad.redhat.com> <6bc2d8e.1e4b.14e0942385e.Coremail.cauchy-love@163.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="bAmEntskrkuBymla" Content-Disposition: inline In-Reply-To: <6bc2d8e.1e4b.14e0942385e.Coremail.cauchy-love@163.com> Subject: Re: [Qemu-devel] Greate difference of disk I/O performance for guest on Qemu-2.30 of CentOS. List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: cauchy-love Cc: qemu-devel@nongnu.org --bAmEntskrkuBymla Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Jun 19, 2015 at 08:40:11AM +0800, cauchy-love wrote: >=20 > the iozone command line is: > ./iozone -i 0 -i 1 -f ./iotmp -Rab ./iotmp.xls -g 8G -n 4G -c >=20 > The problem is why the performance difference is so big for different lin= ux kernels. The guest's io performance test is by both FTP tools and raw wr= ite, which demonstrate the much lower performance of Centos 7 (1/10 of tha= t on Centos 6.5). And how can I debug this problem? You are missing the -I command-line option for O_DIRECT. When you run without -I you are not benchmarking just disk I/O but also page cache performance, which varies depending on the amount of RAM available. This means you cannot compare guest results with host results. Stefan --bAmEntskrkuBymla Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJVhDRKAAoJEJykq7OBq3PICa0H/RyqVHp1FLokqx+uFeyScLTo AouaOCjBkOsgdvf/XktbsHtsIaQ99Iu6Prg8g/iO6e/rzzPtOQzRu0lnZAcA5twy ZaVmfst++18VEc9ueIo24Ld89JUGEJB6P7TURfWXH0LHcMU/Ucbuc8FgJerlg+kI gFzpKz7B6ZBPPgTmGjd1RR76GLlnJAkl9a3V0jUJM9zSz0yI/j3Ux8wuHVvyl4BS uzWac1J5tqQ8UQpVz9oFO87o0VJq9mMMU6PItnk5XNHw7TUW5jghZwlS0Ensw2wc JQ0VdCdTq+1UniZFaWr3DCDVDMUuALRxZxvfozkIg5mjTf+KWNHke5sdkfDzUE4= =uDZH -----END PGP SIGNATURE----- --bAmEntskrkuBymla--