Re: [Qemu-devel] virtio-scsi vs. virtio-blk

[Paolo Bonzini]

Il 09/08/2012 14:31, Stefan Priebe - Profihost AG ha scritto:
> Am 09.08.2012 14:19, schrieb Paolo Bonzini:
>> Il 09/08/2012 14:08, Stefan Priebe - Profihost AG ha scritto:
>>>
>>> virtio-scsi:
>>> rand 4k:
>>>    write: io=822448KB, bw=82228KB/s, iops=20557, runt= 10002msec
>>>    read : io=950920KB, bw=94694KB/s, iops=23673, runt= 10042msec
>>> seq:
>>>    write: io=2436MB, bw=231312KB/s, iops=56, runt= 10784msec
>>>    read : io=3248MB, bw=313799KB/s, iops=76, runt= 10599msec
>>>
>>> virtio-blk:
>>> rand 4k:
>>>    write: io=896472KB, bw=89051KB/s, iops=22262, runt= 10067msec
>>>    read : io=1710MB, bw=175073KB/s, iops=43768, runt= 10002msec
>>> seq:
>>>    write: io=4008MB, bw=391285KB/s, iops=95, runt= 10489msec
>>>    read : io=5748MB, bw=570178KB/s, iops=139, runt= 10323msec
>>
>> Thanks; some overhead is expected, but not this much.  Especially the
>> sequential case is bad, what disk is this?
> 
> right now this is an external iscsi Nexentastor. Locally i can't get
> this bandwith nor these iops to test.

Thanks for the information.

>> Things to test include:
>>
>> - using the deadline I/O scheduler on at least the host, and possibly
>> the guest too
> guest uses noop right now. Disk Host is nexentastor running open
> solaris. I use libiscsi right now so the disks are not visible in both
> cases (virtio-blk and virtio-scsi) to the host right now.

Ok, try deadline in the guest then.  Using noop amplifies bad
performance, because you lose request merging.  With no host scheduler,
as is the case with libiscsi, noop is rarely a good idea.

>> - running perf on the guest and the host (separately) to get profiles
> Which command of perf? Just perf top?

Yeah, perhaps you could make the benchmarks a bit longer so that perf
top has more time to stabilize.  Then you can just cut-and-paste perf
top output.

Paolo