linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Ingo Molnar <mingo@kernel.org>
To: Michael Wang <wangyun@linux.vnet.ibm.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Paul Turner <pjt@google.com>, Mike Galbraith <efault@gmx.de>,
	Andrew Morton <akpm@linux-foundation.org>,
	alex.shi@intel.com, Ram Pai <linuxram@us.ibm.com>,
	"Nikunj A. Dadhania" <nikunj@linux.vnet.ibm.com>,
	Namhyung Kim <namhyung@kernel.org>
Subject: Re: [RFC PATCH v3 0/3] sched: simplify the select_task_rq_fair()
Date: Wed, 20 Feb 2013 11:49:58 +0100	[thread overview]
Message-ID: <20130220104958.GA9152@gmail.com> (raw)
In-Reply-To: <51079178.3070002@linux.vnet.ibm.com>


* Michael Wang <wangyun@linux.vnet.ibm.com> wrote:

> v3 change log:
> 	Fix small logical issues (Thanks to Mike Galbraith).
> 	Change the way of handling WAKE.
> 
> This patch set is trying to simplify the select_task_rq_fair() 
> with schedule balance map.
> 
> After get rid of the complex code and reorganize the logical, 
> pgbench show the improvement, more the clients, bigger the 
> improvement.
> 
> 				Prev:		Post:
> 
> 	| db_size | clients |	|  tps  |	|  tps  |
>         +---------+---------+   +-------+       +-------+
>         | 22 MB   |       1 |   | 10788 |       | 10881 |
>         | 22 MB   |       2 |   | 21617 |       | 21837 |
>         | 22 MB   |       4 |   | 41597 |       | 42645 |
>         | 22 MB   |       8 |   | 54622 |       | 57808 |
>         | 22 MB   |      12 |   | 50753 |       | 54527 |
>         | 22 MB   |      16 |   | 50433 |       | 56368 |	+11.77%
>         | 22 MB   |      24 |   | 46725 |       | 54319 |	+16.25%
>         | 22 MB   |      32 |   | 43498 |       | 54650 |	+25.64%
>         | 7484 MB |       1 |   |  7894 |       |  8301 |
>         | 7484 MB |       2 |   | 19477 |       | 19622 |
>         | 7484 MB |       4 |   | 36458 |       | 38242 |
>         | 7484 MB |       8 |   | 48423 |       | 50796 |
>         | 7484 MB |      12 |   | 46042 |       | 49938 |
>         | 7484 MB |      16 |   | 46274 |       | 50507 |	+9.15%
>         | 7484 MB |      24 |   | 42583 |       | 49175 |	+15.48%
>         | 7484 MB |      32 |   | 36413 |       | 49148 |	+34.97%
>         | 15 GB   |       1 |   |  7742 |       |  7876 |
>         | 15 GB   |       2 |   | 19339 |       | 19531 |
>         | 15 GB   |       4 |   | 36072 |       | 37389 |
>         | 15 GB   |       8 |   | 48549 |       | 50570 |
>         | 15 GB   |      12 |   | 45716 |       | 49542 |
>         | 15 GB   |      16 |   | 46127 |       | 49647 |	+7.63%
>         | 15 GB   |      24 |   | 42539 |       | 48639 |	+14.34%
>         | 15 GB   |      32 |   | 36038 |       | 48560 |	+34.75%
> 
> Please check the patch for more details about schedule balance map.

The changes look clean and reasoable, any ideas exactly *why* it 
speeds up?

I.e. are there one or two key changes in the before/after logic 
and scheduling patterns that you can identify as causing the 
speedup?

Such changes also typically have a chance to cause regressions 
in other workloads - when that happens we need this kind of 
information to be able to enact plan-B.

Thanks,

	Ingo

  parent reply	other threads:[~2013-02-20 10:50 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-01-29  9:08 [RFC PATCH v3 0/3] sched: simplify the select_task_rq_fair() Michael Wang
2013-01-29  9:09 ` [RFC PATCH v3 1/3] sched: schedule balance map foundation Michael Wang
2013-02-20 13:21   ` Peter Zijlstra
2013-02-21  4:52     ` Michael Wang
2013-02-20 13:25   ` Peter Zijlstra
2013-02-21  4:58     ` Michael Wang
2013-02-21 11:37       ` Peter Zijlstra
2013-02-22  2:53         ` Michael Wang
2013-02-22  3:33           ` Alex Shi
2013-02-22  4:19             ` Michael Wang
2013-02-22  4:46               ` Alex Shi
2013-02-22  5:05                 ` Michael Wang
2013-01-29  9:09 ` [RFC PATCH v3 2/3] sched: build schedule balance map Michael Wang
2013-01-29  9:10 ` [RFC PATCH v3 3/3] sched: simplify select_task_rq_fair() with " Michael Wang
2013-02-18  5:52 ` [RFC PATCH v3 0/3] sched: simplify the select_task_rq_fair() Michael Wang
2013-02-20 10:49 ` Ingo Molnar [this message]
2013-02-20 13:32   ` Peter Zijlstra
2013-02-20 14:05     ` Mike Galbraith
2013-02-21  5:21       ` Michael Wang
2013-02-21  5:14     ` Michael Wang
2013-02-21  4:51   ` Michael Wang
2013-02-21  6:11     ` Mike Galbraith
2013-02-21  7:00       ` Michael Wang
2013-02-21  8:10         ` Mike Galbraith
2013-02-21  9:08           ` Michael Wang
2013-02-21  9:43             ` Mike Galbraith
2013-02-22  2:36               ` Michael Wang
2013-02-22  5:02                 ` Mike Galbraith
2013-02-22  5:26                   ` Michael Wang
2013-02-22  6:13                     ` Mike Galbraith
2013-02-22  6:42                   ` Michael Wang
2013-02-22  8:17                     ` Mike Galbraith
2013-02-22  8:35                       ` Michael Wang
2013-02-22  8:21                 ` Peter Zijlstra
2013-02-22  9:10                   ` Michael Wang
2013-02-22  9:39                     ` Peter Zijlstra
2013-02-22  9:58                       ` Michael Wang
2013-02-21  9:20           ` Michael Wang
2013-02-21 10:20     ` Peter Zijlstra
2013-02-22  2:37       ` Michael Wang
2013-02-22  5:08         ` Mike Galbraith
2013-02-22  6:06           ` Michael Wang
2013-02-22  6:19             ` Mike Galbraith
2013-02-22  8:36         ` Peter Zijlstra
2013-02-22  9:11           ` Michael Wang
2013-02-22  9:57             ` Peter Zijlstra
2013-02-22 10:08               ` Michael Wang
2013-02-22  9:40           ` Mike Galbraith
2013-02-22  9:54             ` Ingo Molnar
2013-02-22 10:01               ` Mike Galbraith
2013-02-22 12:11                 ` Ingo Molnar
2013-02-22 12:35                   ` Mike Galbraith
2013-02-22 13:06                     ` Ingo Molnar
2013-02-22 14:30                       ` Mike Galbraith
2013-02-22 14:42                         ` Mike Galbraith

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=20130220104958.GA9152@gmail.com \
    --to=mingo@kernel.org \
    --cc=a.p.zijlstra@chello.nl \
    --cc=akpm@linux-foundation.org \
    --cc=alex.shi@intel.com \
    --cc=efault@gmx.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linuxram@us.ibm.com \
    --cc=namhyung@kernel.org \
    --cc=nikunj@linux.vnet.ibm.com \
    --cc=pjt@google.com \
    --cc=wangyun@linux.vnet.ibm.com \
    /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).