Re: [Qemu-devel] [RFC]QEMU disk I/O limits

[Mike Snitzer]

On Tue, May 31 2011 at  2:39pm -0400,
Anthony Liguori <address@hidden> wrote:

> On 05/31/2011 12:59 PM, Vivek Goyal wrote:
> >On Tue, May 31, 2011 at 09:25:31AM -0500, Anthony Liguori wrote:
> >>On 05/31/2011 09:04 AM, Vivek Goyal wrote:
> >>>On Tue, May 31, 2011 at 08:50:40AM -0500, Anthony Liguori wrote:
> >>>>On 05/31/2011 08:45 AM, Vivek Goyal wrote:
> >>>>>On Mon, May 30, 2011 at 01:09:23PM +0800, Zhi Yong Wu wrote:
> >>>>>>Hello, all,
> >>>>>>
> >>>>>>     I have prepared to work on a feature called "Disk I/O limits" for 
> >>>>>> qemu-kvm projeect.
> >>>>>>     This feature will enable the user to cap disk I/O amount
> performed by a VM.It is important for some storage resources to be
> shared among multi-VMs. As you've known, if some of VMs are doing
> excessive disk I/O, they will hurt the performance of other VMs. 
> >>>>>>
> >>>>>
> >>>>>Hi Zhiyong,
> >>>>>
> >>>>>Why not use kernel blkio controller for this and why reinvent the wheel
> >>>>>and implement the feature again in qemu?
> >>>>
> >>>>blkio controller only works for block devices.  It doesn't work when
> >>>>using files.
> >>>
> >>>So can't we comeup with something to easily determine which device backs
> >>>up this file? Though that will still not work for NFS backed storage
> >>>though.
> >>
> >>Right.
> >>
> >>Additionally, in QEMU, we can rate limit based on concepts that make
> >>sense to a guest.  We can limit the actual I/O ops visible to the
> >>guest which means that we'll get consistent performance regardless
> >>of whether the backing file is qcow2, raw, LVM, or raw over NFS.
> >>
> >
> >Are you referring to merging taking place which can change the definition
> >of IOPS as seen by guest?
> 
> No, with qcow2, it may take multiple real IOPs for what the guest
> sees as an IOP.
> 
> That's really the main argument I'm making here.  The only entity
> that knows what a guest IOP corresponds to is QEMU.  On the backend,
> it may end up being a network request, multiple BIOs to physical
> disks, file access, etc.

Couldn't QEMU give a hint to the kernel about the ratio of guest IOP to
real IOPs?  Or is QEMU blind to the real IOPs that correspond to a guest
IOP?

If QEMU is trully blind to the amount of real IOPs then couldn't QEMU
driven throttling cause physical resources to be oversubscribed
(underestimating the backend work it is creating)?

Mike