From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrea Righi Subject: Re: dm-ioband + bio-cgroup benchmarks Date: Fri, 19 Sep 2008 22:28:40 +0200 Message-ID: <48D40B78.6060709@gmail.com> References: <20080918.210418.226794540.ryov@valinux.co.jp> <20080918131554.GB20640@redhat.com> <20080919.202031.86647893.taka@valinux.co.jp> <20080919131019.GA3606@redhat.com> Reply-To: righi.andrea-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20080919131019.GA3606-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org Errors-To: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org To: Vivek Goyal Cc: xen-devel-GuqFBffKawuULHF6PoxzQEEOCMrvLtNR@public.gmane.org, containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, jens.axboe-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, virtualization-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, dm-devel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org, agk-9JcytcrH/bA+uJoB2kUjGw@public.gmane.org, xemul-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org, fernando-gVGce1chcLdL9jVzuh4AOg@public.gmane.org, balbir-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org List-Id: dm-devel.ids Vivek Goyal wrote: > On Fri, Sep 19, 2008 at 08:20:31PM +0900, Hirokazu Takahashi wrote: >>> To avoid creation of stacking another device (dm-ioband) on top of every >>> device we want to subject to rules, I was thinking of maintaining an >>> rb-tree per request queue. Requests will first go into this rb-tree upon >>> __make_request() and then will filter down to elevator associated with the >>> queue (if there is one). This will provide us the control of releasing >>> bio's to elevaor based on policies (proportional weight, max bandwidth >>> etc) and no need of stacking additional block device. >> I think it's a bit late to control I/O requests there, since process >> may be blocked in get_request_wait when the I/O load is high. >> Please imagine the situation that cgroups with low bandwidths are >> consuming most of "struct request"s while another cgroup with a high >> bandwidth is blocked and can't get enough "struct request"s. >> >> It means cgroups that issues lot of I/O request can win the game. >> > > Ok, this is a good point. Because number of struct requests are limited > and they seem to be allocated on first come first serve basis, so if a > cgroup is generating lot of IO, then it might win. > > But dm-ioband will face the same issue. Essentially it is also a request > queue and it will have limited number of request descriptors. Have you > modified the logic somewhere for allocation of request descriptors to the > waiting processes based on their weights? If yes, the logic probably can > be implemented here too. Maybe throttling dirty page ratio in memory could help to avoid this problem. I mean, if a cgroup is exceeding the i/o limits do ehm... something.. also at the balance_dirty_pages() level. -Andrea