Hi,

I've been following the evolution of this patch with great interest for use
in our qemu-kvm based IaaS public cloud.

I am not a qemu developer, but have watched this patch go through many
rounds of review and we are very much hoping that it makes it into QEMU 1.0.

In a multi-customer multi-VM public cloud environment, disk i/o limiting is
absolutely key, and qemu itself is a much better place to do it than with a
mix of cgroups and other technologies for each different storage backend.
I understand that it is also more efficient:

  
http://www.linux-kvm.org/wiki/images/7/72/2011-forum-keep-a-limit-on-it-io-throttling-in-qemu.pdf

As I say, I am not a qemu developer, so cannot practically help, but wanted
to post to emphasize the importance of this technology as a consumer of qemu.

Best regards,

Richard.

Reply via email to