linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrea Arcangeli <andrea@suse.de>
To: Andrew Morton <akpm@osdl.org>
Cc: wli@holomorphy.com, kernel@kolivas.org,
	chris@cvine.freeserve.co.uk, riel@redhat.com,
	linux-kernel@vger.kernel.org, mbligh@aracnet.com
Subject: Re: 2.6.0-test9 - poor swap performance on low end machines
Date: Tue, 16 Dec 2003 00:37:46 +0100	[thread overview]
Message-ID: <20031215233746.GO6730@dualathlon.random> (raw)
In-Reply-To: <20031215153122.1d915475.akpm@osdl.org>

On Mon, Dec 15, 2003 at 03:31:22PM -0800, Andrew Morton wrote:
> Single-threaded qsbench is OK on 2.6.  Last time I looked it was a little
> quicker than 2.4.  It's when you go to multiple qsbench instances that
> everything goes to crap.
> 
> It's interesting to watch the `top' output during the run.  In 2.4 you see
> three qsbench instances have consumed 0.1 seconds CPU and the fourth has
> consumed 45 seconds and then exits.
> 
> In 2.6 all four processes consume CPU at the same rate.  Really, really
> slowly.

sounds good, so this seems only a fariness issue. 2.6 is more fair but
fariness in this case means much inferior performance.

The reason 2.4 runs faster could be a more aggressive "young" pagetable
heuristic via the swap_out clock algorithm. as soon as one program grows
a bit its rss, it will run for longer, and the longer it runs the more
pages it marks "young" during a clock scan, and the more pages it marks
young the bigger it will grow. This keeps going until it's the by far
biggest task and takes almost all available cpu. This is optimal for
performance, but not optimal for fariness. So 2.6 may be better or worse
depending if fariness payoffs or not, obviously in qsbench it doesn't
since it's not even measured.

  reply	other threads:[~2003-12-15 23:36 UTC|newest]

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-10-29 22:30 2.6.0-test9 - poor swap performance on low end machines Chris Vine
2003-10-31  3:57 ` Rik van Riel
2003-10-31 11:26   ` Roger Luethi
2003-10-31 12:37     ` Con Kolivas
2003-10-31 12:59       ` Roger Luethi
2003-10-31 12:55     ` Ed Tomlinson
2003-11-01 18:34       ` Pasi Savolainen
2003-11-06 18:40       ` bill davidsen
2003-10-31 21:52   ` Chris Vine
2003-11-02 23:06   ` Chris Vine
2003-11-03  0:48     ` Con Kolivas
2003-11-03 21:13       ` Chris Vine
2003-11-04  2:55         ` Con Kolivas
2003-11-04 22:08           ` Chris Vine
2003-11-04 22:30             ` Con Kolivas
2003-12-08 13:52           ` William Lee Irwin III
2003-12-08 14:23             ` Con Kolivas
2003-12-08 14:30               ` William Lee Irwin III
2003-12-09 21:03               ` Chris Vine
2003-12-13 14:08               ` Chris Vine
2003-12-08 19:49             ` Roger Luethi
2003-12-08 20:48               ` William Lee Irwin III
2003-12-09  0:27                 ` Roger Luethi
2003-12-09  4:05                   ` William Lee Irwin III
2003-12-09 15:11                     ` Roger Luethi
2003-12-09 16:04                       ` Rik van Riel
2003-12-09 16:31                         ` Roger Luethi
2003-12-09 18:31                       ` William Lee Irwin III
2003-12-09 19:38                       ` William Lee Irwin III
2003-12-10 13:58                         ` Roger Luethi
2003-12-10 17:47                           ` William Lee Irwin III
2003-12-10 22:23                             ` Roger Luethi
2003-12-11  0:12                               ` William Lee Irwin III
2003-12-10 21:04                           ` Rik van Riel
2003-12-10 23:17                             ` Roger Luethi
2003-12-11  1:31                               ` Rik van Riel
2003-12-11 10:16                                 ` Roger Luethi
2003-12-10 23:30                           ` Helge Hafting
2003-12-10 21:52                 ` Andrea Arcangeli
2003-12-10 22:05                   ` Roger Luethi
2003-12-10 22:44                     ` Andrea Arcangeli
2003-12-11  1:28                       ` William Lee Irwin III
2003-12-11  1:32                         ` Rik van Riel
2003-12-11 10:16                       ` Roger Luethi
2003-12-15 23:31                       ` Andrew Morton
2003-12-15 23:37                         ` Andrea Arcangeli [this message]
2003-12-15 23:54                           ` Andrew Morton
2003-12-16  0:17                             ` Rik van Riel
2003-12-16 11:23                             ` Roger Luethi
2003-12-16 16:29                               ` Rik van Riel
2003-12-17 11:03                                 ` Roger Luethi
2003-12-17 11:06                                   ` William Lee Irwin III
2003-12-17 16:50                                     ` Roger Luethi
2003-12-17 11:33                                   ` Rik van Riel
2003-12-17 18:53                               ` Rik van Riel
2003-12-17 19:27                                 ` William Lee Irwin III
2003-12-17 19:51                                   ` Rik van Riel
2003-12-17 19:49                                 ` Roger Luethi
2003-12-17 21:41                                   ` Andrew Morton
2003-12-17 21:41                                   ` Roger Luethi
2003-12-18  0:21                                     ` Rik van Riel
2003-12-18 22:53                                       ` Roger Luethi
2003-12-18 23:38                                         ` William Lee Irwin III

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=20031215233746.GO6730@dualathlon.random \
    --to=andrea@suse.de \
    --cc=akpm@osdl.org \
    --cc=chris@cvine.freeserve.co.uk \
    --cc=kernel@kolivas.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mbligh@aracnet.com \
    --cc=riel@redhat.com \
    --cc=wli@holomorphy.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).