linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
To: Paul Turner <pjt@google.com>
Cc: linux-kernel@vger.kernel.org,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Bharata B Rao <bharata@linux.vnet.ibm.com>,
	Dhaval Giani <dhaval.giani@gmail.com>,
	Balbir Singh <balbir@linux.vnet.ibm.com>,
	Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>,
	Srivatsa Vaddagiri <vatsa@in.ibm.com>,
	Kamalesh Babulal <kamalesh@linux.vnet.ibm.com>,
	Ingo Molnar <mingo@elte.hu>, Pavel Emelyanov <xemul@openvz.org>
Subject: Re: [patch 15/16] sched: return unused runtime on voluntary sleep
Date: Wed, 22 Jun 2011 18:39:53 +0900	[thread overview]
Message-ID: <4E01B869.1090307@jp.fujitsu.com> (raw)
In-Reply-To: <20110621071701.165027089@google.com>

(2011/06/21 16:17), Paul Turner wrote:
> When a local cfs_rq blocks we return the majority of its remaining quota to the
> global bandwidth pool for use by other runqueues.
> 
> We do this only when the quota is current and there is more than 
> min_cfs_rq_quota [1ms by default] of runtime remaining on the rq.
> 
> In the case where there are throttled runqueues and we have sufficient
> bandwidth to meter out a slice, a second timer is kicked off to handle this
> delivery, unthrottling where appropriate.
> 
> Using a 'worst case' antagonist which executes on each cpu
> for 1ms before moving onto the next on a fairly large machine:
> 
> no quota generations:
>  197.47 ms       /cgroup/a/cpuacct.usage
>  199.46 ms       /cgroup/a/cpuacct.usage
>  205.46 ms       /cgroup/a/cpuacct.usage
>  198.46 ms       /cgroup/a/cpuacct.usage
>  208.39 ms       /cgroup/a/cpuacct.usage
> Since we are allowed to use "stale" quota our usage is effectively bounded by
> the rate of input into the global pool and performance is relatively stable.
> 
> with quota generations [1s increments]:
>  119.58 ms       /cgroup/a/cpuacct.usage
>  119.65 ms       /cgroup/a/cpuacct.usage
>  119.64 ms       /cgroup/a/cpuacct.usage
>  119.63 ms       /cgroup/a/cpuacct.usage
>  119.60 ms       /cgroup/a/cpuacct.usage
> The large deficit here is due to quota generations (/intentionally/) preventing
> us from now using previously stranded slack quota.  The cost is that this quota
> becomes unavailable.
> 
> with quota generations and quota return:
>  200.09 ms       /cgroup/a/cpuacct.usage
>  200.09 ms       /cgroup/a/cpuacct.usage
>  198.09 ms       /cgroup/a/cpuacct.usage
>  200.09 ms       /cgroup/a/cpuacct.usage
>  200.06 ms       /cgroup/a/cpuacct.usage
> By returning unused quota we're able to both stably consume our desired quota
> and prevent unintentional overages due to the abuse of slack quota from 
> previous quota periods (especially on a large machine).
> 
> Signed-off-by: Paul Turner <pjt@google.com>
> 
> ---

(For all but the patch title:)
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>


Thanks,
H.Seto


  parent reply	other threads:[~2011-06-22  9:40 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-21  7:16 [patch 00/16] CFS Bandwidth Control v7 Paul Turner
2011-06-21  7:16 ` [patch 01/16] sched: (fixlet) dont update shares twice on on_rq parent Paul Turner
2011-06-21  7:16 ` [patch 02/16] sched: hierarchical task accounting for SCHED_OTHER Paul Turner
2011-06-21  7:16 ` [patch 03/16] sched: introduce primitives to account for CFS bandwidth tracking Paul Turner
2011-06-22 10:52   ` Peter Zijlstra
2011-07-06 21:38     ` Paul Turner
2011-07-07 11:32       ` Peter Zijlstra
2011-06-21  7:16 ` [patch 04/16] sched: validate CFS quota hierarchies Paul Turner
2011-06-22  5:43   ` Bharata B Rao
2011-06-22  6:57     ` Paul Turner
2011-06-22  9:38   ` Hidetoshi Seto
2011-06-21  7:16 ` [patch 05/16] sched: accumulate per-cfs_rq cpu usage and charge against bandwidth Paul Turner
2011-06-21  7:16 ` [patch 06/16] sched: add a timer to handle CFS bandwidth refresh Paul Turner
2011-06-22  9:38   ` Hidetoshi Seto
2011-06-21  7:16 ` [patch 07/16] sched: expire invalid runtime Paul Turner
2011-06-22  9:38   ` Hidetoshi Seto
2011-06-22 15:47   ` Peter Zijlstra
2011-06-28  4:42     ` Paul Turner
2011-06-29  2:29       ` Paul Turner
2011-06-21  7:16 ` [patch 08/16] sched: throttle cfs_rq entities which exceed their local runtime Paul Turner
2011-06-22  7:11   ` Bharata B Rao
2011-06-22 16:07   ` Peter Zijlstra
2011-06-22 16:54     ` Paul Turner
2011-06-21  7:16 ` [patch 09/16] sched: unthrottle cfs_rq(s) who ran out of quota at period refresh Paul Turner
2011-06-22 17:29   ` Peter Zijlstra
2011-06-28  4:40     ` Paul Turner
2011-06-28  9:11       ` Peter Zijlstra
2011-06-29  3:37         ` Paul Turner
2011-06-21  7:16 ` [patch 10/16] sched: throttle entities exceeding their allowed bandwidth Paul Turner
2011-06-22  9:39   ` Hidetoshi Seto
2011-06-21  7:17 ` [patch 11/16] sched: allow for positional tg_tree walks Paul Turner
2011-06-21  7:17 ` [patch 12/16] sched: prevent interactions with throttled entities Paul Turner
2011-06-22 21:34   ` Peter Zijlstra
2011-06-28  4:43     ` Paul Turner
2011-06-23 11:49   ` Peter Zijlstra
2011-06-28  4:38     ` Paul Turner
2011-06-21  7:17 ` [patch 13/16] sched: migrate throttled tasks on HOTPLUG Paul Turner
2011-06-21  7:17 ` [patch 14/16] sched: add exports tracking cfs bandwidth control statistics Paul Turner
2011-06-21  7:17 ` [patch 15/16] sched: return unused runtime on voluntary sleep Paul Turner
2011-06-21  7:33   ` Paul Turner
2011-06-22  9:39   ` Hidetoshi Seto [this message]
2011-06-23 15:26   ` Peter Zijlstra
2011-06-28  1:42     ` Paul Turner
2011-06-28 10:01       ` Peter Zijlstra
2011-06-28 18:45         ` Paul Turner
2011-06-21  7:17 ` [patch 16/16] sched: add documentation for bandwidth control Paul Turner
2011-06-21 10:30   ` Hidetoshi Seto
2011-06-21 19:46     ` Paul Turner
2011-06-22 10:05 ` [patch 00/16] CFS Bandwidth Control v7 Hidetoshi Seto
2011-06-23 12:06   ` Peter Zijlstra
2011-06-23 12:43     ` Ingo Molnar
2011-06-24  5:11       ` Hidetoshi Seto
2011-06-26 10:35         ` Ingo Molnar
2011-06-29  4:05           ` Hu Tao
2011-07-01 12:28             ` Ingo Molnar
2011-07-05  3:58               ` Hu Tao
2011-07-05  8:50                 ` Ingo Molnar
2011-07-05  8:52                   ` Ingo Molnar
2011-07-07  3:53                     ` Hu Tao

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4E01B869.1090307@jp.fujitsu.com \
    --to=seto.hidetoshi@jp.fujitsu.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=balbir@linux.vnet.ibm.com \
    --cc=bharata@linux.vnet.ibm.com \
    --cc=dhaval.giani@gmail.com \
    --cc=kamalesh@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=pjt@google.com \
    --cc=svaidy@linux.vnet.ibm.com \
    --cc=vatsa@in.ibm.com \
    --cc=xemul@openvz.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).