[PATCH 18/23] io-controller: blkio_cgroup patches from Ryo to track async bios.
balbir at linux.vnet.ibm.com
Tue Sep 1 20:12:34 PDT 2009
On Wed, Sep 2, 2009 at 6:29 AM, KAMEZAWA
Hiroyuki<kamezawa.hiroyu at jp.fujitsu.com> wrote:
> On Tue, 1 Sep 2009 10:11:42 -0400
> Vivek Goyal <vgoyal at redhat.com> wrote:
>> > > > - Somebody also gave an example where there is a memory hogging process and
>> > > > possibly pushes out some processes to swap. It does not sound fair to
>> > > > charge those proccess for that swap writeout. These processes never
>> > > > requested swap IO.
>> > I think that swap writeouts should be charged to the memory hogging
>> > process, because the process consumes more resources and it should get
>> > a penalty.
>> A process requesting memory gets IO penalty? IMHO, swapping is a kernel
>> mechanism and kernel's way of providing extended RAM. If we want to solve
>> the issue of memory hogging by a process then right way to solve is to use
>> memory controller and not by charging the process for IO activity.
>> Instead, proabably a more suitable way is to charge swap activity to root
>> group (where by default all the kernel related activity goes).
> I agree. It't memcg's job.
> (Support dirty_ratio in memcg is necessary, I think)
> background-write-out-to-swap-for-memory-shortage should be handled
> as kernel I/O. If swap-out-by-memcg bacause of its limit is a problem,
> dirty_ratio for memcg should be implemetned.
I tend to agree, looks like dirty_ratio will become important along
with overcommit support in the future.
More information about the Containers