Re: [ceph-users] Increase queue_depth in KVM
Konstantin, Thanks for explanation. But unfortunately, upgrading qemu is nearly impossible in my case. So is there something else I can do, or I have to agree with fact that write IOPS had to be 8x smaller inside KVM rather than outside KVM? :| pt., 13 lip 2018 o 04:22 Konstantin Shalygin napisał(a): > > I've seen some people using 'num_queues' but I don't have this parameter > > in my schemas(libvirt version = 1.3.1, qemu version = 2.5.0 > > > num-queues is available from qemu 2.7 [1] > > > [1] https://wiki.qemu.org/ChangeLog/2.7 > > > > > k > > ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Increase queue_depth in KVM
I've seen some people using 'num_queues' but I don't have this parameter in my schemas(libvirt version = 1.3.1, qemu version = 2.5.0 num-queues is available from qemu 2.7 [1] [1] https://wiki.qemu.org/ChangeLog/2.7 k ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Increase queue_depth in KVM
Hello, Steffen, Thanks for Your reply. Sorry but I was on holidays, now I'm back and still digging into my problem.. :( I've read thousands of google links but can't find anything which could help me. - tried all qemu drive IO(io=) and cache(cache=) modes, nothing could come even close to the results I'm getting outside KVM. - enabling blk-mq inside KVM guest didn't help - enabling iothreads didn't help - 'queues' parameter in my libvirt schemas can be only applied to 'virtio-serial', can't use it in virtio-scsi nor virtio-blk) - I've seen some people using 'num_queues' but I don't have this parameter in my schemas(libvirt version = 1.3.1, qemu version = 2.5.0) So, is there really no way to increase queue_depth to rbd device in kvm domain or any other way to achieve results similar to those obtained outside KVM? :/ wt., 26 cze 2018 o 15:19 Steffen Winther Sørensen napisał(a): > > > > On 26 Jun 2018, at 14.04, Damian Dabrowski wrote: > > > > Hi Stefan, thanks for reply. > > > > Unfortunately it didn't work. > > > > disk config: > > > >discard='unmap'/> > > > > > > > >name='volumes-nvme/volume-ce247187-a625-49f1-bacd-fc03df215395'> > > > > > > > > > > > > ce247187-a625-49f1-bacd-fc03df215395 > > > > > > > > > > Controller config: > > > > > >function='0x0'/> > > > > > > > > benchmark command: fio --randrepeat=1 --ioengine=libaio --direct=1 > --name=test --filename=test --bs=4k --iodepth=64 --size=1G > --readwrite=randwrite --time_based --runtime=60 > --write_iops_log=write_results --numjobs=8 > > > > And I'm still getting very low random write IOPS inside KVM instance > with 8vcores(3-5k compared to 20k+ outside KVM) > > > > Maybe do You have any idea how to deal with it? > What about trying with io=‘threads’ and/or maybe cache=’none’ or swap from > virtio-scsi to blk-mq? > > Other people had similar issues, try asking ‘G’ > > https://serverfault.com/questions/425607/kvm-guest-io-is-much-slower-than-host-io-is-that-normal > https://wiki.mikejung.biz/KVM_/_Xen > > /Steffen > > ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Increase queue_depth in KVM
> On 26 Jun 2018, at 14.04, Damian Dabrowski wrote: > > Hi Stefan, thanks for reply. > > Unfortunately it didn't work. > > disk config: > >discard='unmap'/> > > > >name='volumes-nvme/volume-ce247187-a625-49f1-bacd-fc03df215395'> > > > > > > ce247187-a625-49f1-bacd-fc03df215395 > > > > > Controller config: > > >function='0x0'/> > > > > benchmark command: fio --randrepeat=1 --ioengine=libaio --direct=1 > --name=test --filename=test --bs=4k --iodepth=64 --size=1G > --readwrite=randwrite --time_based --runtime=60 > --write_iops_log=write_results --numjobs=8 > > And I'm still getting very low random write IOPS inside KVM instance with > 8vcores(3-5k compared to 20k+ outside KVM) > > Maybe do You have any idea how to deal with it? What about trying with io=‘threads’ and/or maybe cache=’none’ or swap from virtio-scsi to blk-mq? Other people had similar issues, try asking ‘G’ https://serverfault.com/questions/425607/kvm-guest-io-is-much-slower-than-host-io-is-that-normal https://wiki.mikejung.biz/KVM_/_Xen /Steffen ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Increase queue_depth in KVM
Hi Stefan, thanks for reply. Unfortunately it didn't work. disk config: ce247187-a625-49f1-bacd-fc03df215395 Controller config: benchmark command: fio --randrepeat=1 --ioengine=libaio --direct=1 --name=test --filename=test --bs=4k --iodepth=64 --size=1G --readwrite=randwrite --time_based --runtime=60 --write_iops_log=write_results --numjobs=8 And I'm still getting very low random write IOPS inside KVM instance with 8vcores(3-5k compared to 20k+ outside KVM) Maybe do You have any idea how to deal with it? wt., 26 cze 2018 o 09:37 Stefan Kooman napisał(a): > Quoting Damian Dabrowski (scoot...@gmail.com): > > Hello, > > > > When I mount rbd image with -o queue_depth=1024 I can see much > improvement, > > generally on writes(random write improvement from 3k IOPS on standard > > queue_depth to 24k IOPS on queue_depth=1024). > > > > But is there any way to attach rbd disk to KVM instance with custom > > queue_depth? I can't find any information about it. > > Not specifically "queue depth", but if you use virtio-scsi, and have a > VM with more than 1 vCPU, you can give each vCPU it's own queue [1]: > > > > > > > Gr. Stefan > > [1]: > https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/7/html/virtualization_tuning_and_optimization_guide/sect-virtualization_tuning_optimization_guide-blockio-techniques > > > -- > | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 > | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl > ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Increase queue_depth in KVM
Quoting Damian Dabrowski (scoot...@gmail.com): > Hello, > > When I mount rbd image with -o queue_depth=1024 I can see much improvement, > generally on writes(random write improvement from 3k IOPS on standard > queue_depth to 24k IOPS on queue_depth=1024). > > But is there any way to attach rbd disk to KVM instance with custom > queue_depth? I can't find any information about it. Not specifically "queue depth", but if you use virtio-scsi, and have a VM with more than 1 vCPU, you can give each vCPU it's own queue [1]: Gr. Stefan [1]: https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/7/html/virtualization_tuning_and_optimization_guide/sect-virtualization_tuning_optimization_guide-blockio-techniques -- | BIT BV http://www.bit.nl/Kamer van Koophandel 09090351 | GPG: 0xD14839C6 +31 318 648 688 / i...@bit.nl ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] Increase queue_depth in KVM
Hello, When I mount rbd image with -o queue_depth=1024 I can see much improvement, generally on writes(random write improvement from 3k IOPS on standard queue_depth to 24k IOPS on queue_depth=1024). But is there any way to attach rbd disk to KVM instance with custom queue_depth? I can't find any information about it. Thanks for any informations. ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com