From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:57218) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1S5Brg-0002or-8L for qemu-devel@nongnu.org; Wed, 07 Mar 2012 03:04:53 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1S5BrD-0002DR-Au for qemu-devel@nongnu.org; Wed, 07 Mar 2012 03:04:51 -0500 Received: from mail-lpp01m010-f45.google.com ([209.85.215.45]:35128) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1S5BrD-0002DC-12 for qemu-devel@nongnu.org; Wed, 07 Mar 2012 03:04:23 -0500 Received: by lahe6 with SMTP id e6so7737352lah.4 for ; Wed, 07 Mar 2012 00:04:20 -0800 (PST) MIME-Version: 1.0 In-Reply-To: <4F568AAC.6060206@shiftmail.org> References: <20120210143639.GA17883@gmail.com> <4F54E620.8060400@tuxadero.com> <4F54ED84.7030601@tuxadero.com> <4F568AAC.6060206@shiftmail.org> Date: Wed, 7 Mar 2012 08:04:20 +0000 Message-ID: From: Stefan Hajnoczi Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] virtio-blk performance regression and qemu-kvm List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Reeted Cc: Martin Mailand , Dongsu Park , kvm@vger.kernel.org, qemu-devel@nongnu.org On Tue, Mar 6, 2012 at 10:07 PM, Reeted wrote: > On 03/06/12 13:59, Stefan Hajnoczi wrote: >> >> On Mon, Mar 5, 2012 at 4:44 PM, Martin Mailand >> =A0wrote: >>> >>> Am 05.03.2012 17:35, schrieb Stefan Hajnoczi: >>> >>>>> 1. Test on i7 Laptop with Cpu governor "ondemand". >>>>>> >>>>>> =A0v0.14.1 >>>>>> =A0bw=3D63492KB/s iops=3D15873 >>>>>> =A0bw=3D63221KB/s iops=3D15805 >>>>>> >>>>>> =A0v1.0 >>>>>> =A0bw=3D36696KB/s iops=3D9173 >>>>>> =A0bw=3D37404KB/s iops=3D9350 >>>>>> >>>>>> =A0master >>>>>> =A0bw=3D36396KB/s iops=3D9099 >>>>>> =A0bw=3D34182KB/s iops=3D8545 >>>>>> >>>>>> =A0Change the Cpu governor to "performance" >>>>>> =A0master >>>>>> =A0bw=3D81756KB/s iops=3D20393 >>>>>> =A0bw=3D81453KB/s iops=3D20257 >>>> >>>> Interesting finding. =A0Did you show the 0.14.1 results with >>>> "performance" governor? >>> >>> >>> >>> Hi Stefan, >>> all results are with "ondemand" except the one where I changed it to >>> "performance" >>> >>> Do you want a v0.14.1 test with the governor on "performance"? >> >> Yes, the reason why that would be interesting is because it allows us >> to put the performance gain with master+"performance" into >> perspective. =A0We could see how much of a change we get. > > > > Me too, I would be interested in seeing 0.14.1 being tested with performa= nce > governor so to compare it to master with performance governor, to make su= re > that this is not a regression. > > BTW, I'll take the opportunity to say that 15.8 or 20.3 k IOPS are very l= ow > figures compared to what I'd instinctively expect from a paravirtualized > block driver. > There are now PCIe SSD cards that do 240 k IOPS (e.g. "OCZ RevoDrive 3 x2 > max iops") which is 12-15 times higher, for something that has to go thro= ugh > a real driver and a real PCI-express bus, and can't use zero-copy > techniques. > The IOPS we can give to a VM is currently less than half that of a single > SSD SATA drive (60 k IOPS or so, these days). > That's why I consider this topic of virtio-blk performances very importan= t. > I hope there can be improvements in this sector... It depends on the benchmark configuration. virtio-blk is capable of doing 100,000s of iops, I've seen results. My guess is that you can do >100,000 read iops with virtio-blk on a good machine and stock qemu-kvm. Stefan