Re: [patch 1/4] io controller: documentation

[Vivek Goyal <vgoyal@redhat.com>]

On Mon, Nov 10, 2008 at 10:48:54AM +0800, Li Zefan wrote:
> Hi, Vivek
> 
> Thanks for your work. :)
> 
> A question below.
> 
> > +For example, assume there are two cgroups A and B with weights 1024 and 2048
> > +in the system. Tasks in two cgroups A and B are doing IO to two disks sda and
> > +sdb in the system. A user has enabled IO control on both sda and sdb. Now on
> > +both sda and sdb, tasks in cgroup B will get to use 2/3 of disk BW and
> > +tasks in cgroup A will get to use 1/3 of disk bandwidth, only in case of
> > +contention. If tasks in any of the groups stop doing IO to a particular disk,
> > +task in other group will get to use full disk BW for that duration.
> > +
> 
> So in this example, I can't assign 1/3 of sda's disk BW while 2/3 of sdb's
> disk BW to A. Am I right?

Hi,

Yes you are right. Currently policies are global and not per disk. So in 
above example, cgroup A will get 1/3 of disk BW both on sda and sdb. And
one can not configure in such a manner so that A gets 1/3 of BW on sda and
2/3 of BW on sdb.

I think assigning cgroup weights per disk should be doable but personally
I think it makes configuration complex and I am not sure if it really a
very useful feature.

Thanks
Vivek