From: Peter Zijlstra <peterz@infradead.org>
To: Jason Low <jason.low2@hp.com>
Cc: Rik van Riel <riel@redhat.com>, Ingo Molnar <mingo@redhat.com>,
LKML <linux-kernel@vger.kernel.org>,
Mike Galbraith <efault@gmx.de>,
Thomas Gleixner <tglx@linutronix.de>,
Paul Turner <pjt@google.com>, Alex Shi <alex.shi@intel.com>,
Preeti U Murthy <preeti@linux.vnet.ibm.com>,
Vincent Guittot <vincent.guittot@linaro.org>,
Morten Rasmussen <morten.rasmussen@arm.com>,
Namhyung Kim <namhyung@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Kees Cook <keescook@chromium.org>, Mel Gorman <mgorman@suse.de>,
aswin@hp.com, scott.norton@hp.com, chegu_vinod@hp.com
Subject: Re: [RFC] sched: Limit idle_balance() when it is being used too frequently
Date: Thu, 18 Jul 2013 11:32:18 +0200 [thread overview]
Message-ID: <20130718093218.GH27075@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <1374120144.1816.45.camel@j-VirtualBox>
On Wed, Jul 17, 2013 at 09:02:24PM -0700, Jason Low wrote:
> I ran a few AIM7 workloads for the 8 socket HT enabled case and I needed
> to set N to more than 20 in order to get the big performance gains.
>
> One thing that I thought of was to have N be based on how often idle
> balance attempts does not pull task(s).
>
> For example, N can be calculated based on the number of idle balance
> attempts for the CPU since the last "successful" idle balance attempt.
> So if the previous 30 idle balance attempts resulted in no tasks moved,
> then n = 30 / 5. So idle balance gets less time to run as the number of
> unneeded idle balance attempts increases, and thus N will not be set too
> high during situations where idle balancing is "successful" more often.
> Any comments on this idea?
It would be good to get a solid explanation for why we need such high N.
But yes that might work.
next prev parent reply other threads:[~2013-07-18 9:32 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-07-16 19:21 Jason Low
2013-07-16 19:27 ` Rik van Riel
2013-07-16 20:20 ` Peter Zijlstra
2013-07-16 22:48 ` Jason Low
2013-07-17 7:25 ` Peter Zijlstra
2013-07-17 7:48 ` Peter Zijlstra
2013-07-17 8:11 ` Jason Low
2013-07-17 9:39 ` Peter Zijlstra
2013-07-17 15:59 ` Jason Low
2013-07-17 16:18 ` Peter Zijlstra
2013-07-17 17:51 ` Rik van Riel
2013-07-17 18:01 ` Peter Zijlstra
2013-07-17 18:48 ` Jason Low
2013-07-18 4:02 ` Jason Low
2013-07-18 9:32 ` Peter Zijlstra [this message]
2013-07-18 11:59 ` Rik van Riel
2013-07-18 12:15 ` Srikar Dronamraju
2013-07-18 12:35 ` Peter Zijlstra
2013-07-18 13:06 ` Srikar Dronamraju
2013-07-18 19:06 ` Jason Low
2013-07-19 18:37 ` Peter Zijlstra
2013-07-19 19:15 ` Jason Low
2013-07-18 12:12 ` Srikar Dronamraju
2013-07-18 19:03 ` Jason Low
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=20130718093218.GH27075@twins.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=akpm@linux-foundation.org \
--cc=alex.shi@intel.com \
--cc=aswin@hp.com \
--cc=chegu_vinod@hp.com \
--cc=efault@gmx.de \
--cc=jason.low2@hp.com \
--cc=keescook@chromium.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@suse.de \
--cc=mingo@redhat.com \
--cc=morten.rasmussen@arm.com \
--cc=namhyung@kernel.org \
--cc=pjt@google.com \
--cc=preeti@linux.vnet.ibm.com \
--cc=riel@redhat.com \
--cc=scott.norton@hp.com \
--cc=tglx@linutronix.de \
--cc=vincent.guittot@linaro.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
Powered by JetHome