Re: qcow2 slowdown in kvm-78,79
This is probably the change to cache=writethrough. I bet if you set cache=writeback then you'll see this go away. cache=writeback is only slightly faster, like going from 6MB/s to 7MB/s. ignore that remark, I retested with cache=writeback and everything is OK again. I must have misspelled somehow and qemu didn't quit on me or there was something else going on at the time of my tests. Thanks! /Henrik Holst https://witsbits.com -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: qcow2 slowdown in kvm-78,79
On Tue, 18 Nov 2008 15:39:52 +0100, Henrik Holst wrote: The bundled qemu in kvm-78 and kvm-79 slows down disk i/o with qcow2 images by an order of 10. If one got 60MB/s before, one gets around 6MB/s with 78 and 79 (measured with dd) so i'm not the only one facing this problem. - Thomas -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: qcow2 slowdown in kvm-78,79
* Henrik Holst [EMAIL PROTECTED] [2008-11-18 08:42]: The bundled qemu in kvm-78 and kvm-79 slows down disk i/o with qcow2 images by an order of 10. If one got 60MB/s before, one gets around 6MB/s with 78 and 79 (measured with dd) dd command? what's your -drive parameters look like, specifically, what if type? ide, scsi, virtio? -- Ryan Harper Software Engineer; Linux Technology Center IBM Corp., Austin, Tx [EMAIL PROTECTED] -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: qcow2 slowdown in kvm-78,79
Ryan Harper wrote: * Henrik Holst [EMAIL PROTECTED] [2008-11-18 08:42]: The bundled qemu in kvm-78 and kvm-79 slows down disk i/o with qcow2 images by an order of 10. If one got 60MB/s before, one gets around 6MB/s with 78 and 79 (measured with dd) dd command? what's your -drive parameters look like, specifically, what if type? ide, scsi, virtio? This is probably the change to cache=writethrough. I bet if you set cache=writeback then you'll see this go away. qcow needs some love. Regards, Anthony Liguori -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: qcow2 slowdown in kvm-78,79
On Wed, 19 Nov 2008 12:26:41 -0600, Anthony Liguori wrote: Ryan Harper wrote: * Henrik Holst [EMAIL PROTECTED] [2008-11-18 08:42]: The bundled qemu in kvm-78 and kvm-79 slows down disk i/o with qcow2 images by an order of 10. If one got 60MB/s before, one gets around 6MB/s with 78 and 79 (measured with dd) dd command? what's your -drive parameters look like, specifically, what if type? ide, scsi, virtio? This is probably the change to cache=writethrough. I bet if you set cache=writeback then you'll see this go away. this was it. cache=writeback and the throughput goes from 10mb/s to 80/mb/ s with if=scsi thanks. - Thomas -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: qcow2 slowdown in kvm-78,79
Thomas Mueller wrote: On Wed, 19 Nov 2008 12:26:41 -0600, Anthony Liguori wrote: This is probably the change to cache=writethrough. I bet if you set cache=writeback then you'll see this go away. this was it. cache=writeback and the throughput goes from 10mb/s to 80/mb/ s with if=scsi Yeah, we'll most likely switch back to cache=writeback for the default for qcow2 images. qcow2 needs some lovin'. Regards, Anthony Liguori thanks. - Thomas -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: qcow2 slowdown in kvm-78,79
Anthony Liguori wrote: Ryan Harper wrote: * Henrik Holst [EMAIL PROTECTED] [2008-11-18 08:42]: The bundled qemu in kvm-78 and kvm-79 slows down disk i/o with qcow2 images by an order of 10. If one got 60MB/s before, one gets around 6MB/s with 78 and 79 (measured with dd) dd command? what's your -drive parameters look like, specifically, what if type? ide, scsi, virtio? This is probably the change to cache=writethrough. I bet if you set cache=writeback then you'll see this go away. Would this affect 'qemu-img commit'? It took over an hour to commit a fresh install of Vista on a qcow2 virtual disk. (The base image expanded from 3GB to 11GB, which is a lot of committing, though.) -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: qcow2 slowdown in kvm-78,79
Would this affect 'qemu-img commit'? It took over an hour to commit a fresh install of Vista on a qcow2 virtual disk. (The base image expanded from 3GB to 11GB, which is a lot of committing, though.) Yes. Regards, Anthony Liguori -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: qcow2 slowdown in kvm-78,79
ons 2008-11-19 klockan 12:26 -0600 skrev Anthony Liguori: Ryan Harper wrote: * Henrik Holst [EMAIL PROTECTED] [2008-11-18 08:42]: The bundled qemu in kvm-78 and kvm-79 slows down disk i/o with qcow2 images by an order of 10. If one got 60MB/s before, one gets around 6MB/s with 78 and 79 (measured with dd) dd command? dd if=/dev/zero of=/mnt/zero count=10240 and then kill -USR1 pid to get the statistics what's your -drive parameters look like, specifically, what if type? ide, scsi, virtio? It is the same regardless of ide,scsi or virtio. Virtio starts quite high around 70MB/s but quickly goes down to 6MB/s (after roughly 10 seconds). The command line (for virtio) is: qemu-system-x86_64 -M pc -m 771 -smp 1 -vnc :0,tls,x509verify=/opt/kvm/cert \ -k en-us -monitor pty -serial none -usbdevice tablet \ -drive file=/mnt/d0/drive.root,if=virtio,media=disk,boot=on \ -drive file=/mnt/d0/drive2,if=virtio,media=disk,boot=off \ -net nic,macaddr=DE:AF:00:00:10:43,model=virtio \ -net tap,script=/opt/kvm/share/qemu-ifup \ -uuid 6b7d5438-1309-4cf0-ae08-57234d61d473 I have tested with all kinds of -m settings as well (up to 8GiB) without any change. This is probably the change to cache=writethrough. I bet if you set cache=writeback then you'll see this go away. cache=writeback is only slightly faster, like going from 6MB/s to 7MB/s. The only thing that works is using qemu userland from kvm-77. I have not tested this with anything other than qcow2 so I'm guessing that it is related to qcow2 due to the cache change in kvm-78 and your other comments about qcow2 and these changes when issues about degraded qcow2 performance has come up. /Henrik Holst https://witsbits.com -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
qcow2 slowdown in kvm-78,79
The bundled qemu in kvm-78 and kvm-79 slows down disk i/o with qcow2 images by an order of 10. If one got 60MB/s before, one gets around 6MB/s with 78 and 79 (measured with dd) Replacing qemu with the one bundled in kvm-77 solves this regression. I have tested all the cache= options to -drive without success. cache=writeback and cache=off was a little bit better than the default cache=writethrough but not with any useful improvement. /Henrik Holst https://Witsbits.com - Utility Computing -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html