From: Helge Hafting <helgehaf@idb.hist.no>
To: Daniel Phillips <phillips@bonn-fries.net>, linux-kernel@vger.kernel.org
Subject: Re: [resent PATCH] Re: very slow parallel read performance
Date: Mon, 27 Aug 2001 09:08:44 +0200 [thread overview]
Message-ID: <3B89F1FC.40F747D4@idb.hist.no> (raw)
In-Reply-To: <Pine.LNX.4.33L.0108241600410.31410-100000@duckman.distro.conectiva> <20010824222226Z16116-32383+1242@humbolt.nl.linux.org>
Daniel Phillips wrote:
>
> On August 24, 2001 09:02 pm, Rik van Riel wrote:
> > I guess in the long run we should have automatic collapse
> > of the readahead window when we find that readahead window
> > thrashing is going on, in the short term I think it is
> > enough to have the maximum readahead size tunable in /proc,
> > like what is happening in the -ac kernels.
>
> Yes, and the most effective way to detect that the readahead window is too
> high is by keeping a history of recently evicted pages. When we find
> ourselves re-reading pages that were evicted before ever being used we know
> exactly what the problem is.
Counting how much we are reading ahead and comparing with total RAM
(or total cache) might also be an idea. We may then read ahead
a lot for those who runs a handful of processes, and
do smaller readahead for those that runs thousands of processes.
Helge Hafting
next prev parent reply other threads:[~2001-08-27 7:10 UTC|newest]
Thread overview: 123+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-08-23 21:35 very slow parallel read performance Lehmann
2001-08-24 7:35 ` [resent PATCH] " Roger Larsson
2001-08-24 17:43 ` Rik van Riel
2001-08-24 18:28 ` Roger Larsson
2001-08-24 19:02 ` Rik van Riel
2001-08-24 20:37 ` Gérard Roudier
2001-08-24 23:12 ` Rik van Riel
2001-08-25 8:02 ` Gérard Roudier
2001-08-25 9:26 ` Roger Larsson
2001-08-25 11:49 ` Gérard Roudier
2001-08-25 17:56 ` Roger Larsson
2001-08-25 19:13 ` Gérard Roudier
2001-08-25 13:17 ` Rik van Riel
2001-08-24 22:29 ` Daniel Phillips
2001-08-24 23:10 ` Rik van Riel
2001-08-25 0:42 ` Daniel Phillips
2001-08-27 7:08 ` Helge Hafting [this message]
2001-08-27 14:31 ` Daniel Phillips
2001-08-27 14:42 ` Alex Bligh - linux-kernel
2001-08-27 15:14 ` Rik van Riel
2001-08-27 16:04 ` Daniel Phillips
[not found] ` <Pine.LNX.4.33L.0108271213370.5646-100000@imladris.rielhome.cone ctiva>
2001-08-27 19:34 ` Alex Bligh - linux-kernel
2001-08-27 20:03 ` Oliver Neukum
2001-08-27 20:19 ` Alex Bligh - linux-kernel
2001-08-27 21:38 ` Oliver.Neukum
2001-08-27 22:26 ` Alex Bligh - linux-kernel
2001-08-27 21:29 ` Daniel Phillips
2001-08-27 16:02 ` Daniel Phillips
2001-08-27 19:36 ` Alex Bligh - linux-kernel
2001-08-27 20:24 ` Daniel Phillips
2001-08-27 16:55 ` David Lang
2001-08-27 18:54 ` Daniel Phillips
2001-08-27 18:37 ` Oliver Neukum
2001-08-27 19:04 ` Daniel Phillips
2001-08-27 19:43 ` Oliver Neukum
2001-08-27 20:37 ` Daniel Phillips
2001-08-27 22:10 ` Oliver.Neukum
2001-08-27 21:44 ` Linus Torvalds
2001-08-27 22:30 ` Daniel Phillips
2001-08-27 23:00 ` Marcelo Tosatti
2001-08-28 3:10 ` Linus Torvalds
2001-08-27 19:55 ` Richard Gooch
2001-08-27 20:09 ` Oliver Neukum
2001-08-27 21:06 ` Daniel Phillips
2001-08-24 20:18 ` Daniel Phillips
2001-08-24 20:19 ` Rik van Riel
2001-08-24 21:11 ` Daniel Phillips
2001-08-24 23:03 ` Rik van Riel
2001-08-25 0:41 ` Daniel Phillips
2001-08-25 1:34 ` Rik van Riel
2001-08-25 15:49 ` Daniel Phillips
2001-08-25 15:50 ` Rik van Riel
2001-08-25 16:28 ` Lehmann
2001-08-25 16:34 ` Rik van Riel
2001-08-25 16:41 ` Lehmann
2001-08-26 16:55 ` Daniel Phillips
2001-08-26 18:39 ` Rik van Riel
2001-08-26 19:46 ` Daniel Phillips
2001-08-26 19:52 ` Rik van Riel
2001-08-26 20:08 ` Daniel Phillips
2001-08-26 22:33 ` Russell King
2001-08-26 23:24 ` Daniel Phillips
2001-08-26 23:24 ` Russell King
2001-08-27 0:07 ` Rik van Riel
2001-08-27 0:02 ` Rik van Riel
2001-08-27 0:42 ` Daniel Phillips
2001-08-25 16:43 ` Daniel Phillips
2001-08-25 19:15 ` Alan Cox
2001-08-25 19:35 ` Lehmann
2001-08-25 20:52 ` Rik van Riel
2001-08-26 0:46 ` John Stoffel
2001-08-26 1:07 ` Alan Cox
2001-08-26 3:30 ` Rik van Riel
2001-08-26 3:40 ` Rik van Riel
2001-08-26 5:28 ` Daniel Phillips
2001-08-26 1:38 ` Daniel Phillips
2001-08-26 2:49 ` Lehmann
2001-08-26 17:29 ` Daniel Phillips
2001-08-26 17:37 ` Craig I. Hagan
2001-08-26 18:56 ` Rik van Riel
2001-08-26 19:18 ` Lehmann
2001-08-26 21:07 ` Daniel Phillips
2001-08-26 22:12 ` Rik van Riel
2001-08-26 23:24 ` Lehmann
2001-08-26 20:26 ` Gérard Roudier
2001-08-26 21:20 ` Daniel Phillips
2001-08-26 3:32 ` Rik van Riel
2001-08-26 13:22 ` Lehmann
2001-08-26 13:48 ` Rik van Riel
2001-08-26 14:55 ` Lehmann
2001-08-26 15:06 ` Rik van Riel
2001-08-26 15:25 ` Lehmann
2001-08-25 21:33 ` Alan Cox
2001-08-25 23:34 ` Lehmann
2001-08-26 2:02 ` Rik van Riel
2001-08-26 2:57 ` Lehmann
2001-08-24 23:23 ` Lehmann
[not found] ` <200108242344.f7ONi0h21270@mailg.telia.com>
2001-08-25 0:28 ` Lehmann
2001-08-25 3:09 ` Rik van Riel
2001-08-25 9:13 ` Gérard Roudier
2001-08-26 16:54 ` Daniel Phillips
2001-08-26 18:59 ` Victor Yodaiken
2001-08-26 19:38 ` Rik van Riel
2001-08-26 20:05 ` Victor Yodaiken
2001-08-26 20:34 ` Rik van Riel
2001-08-26 20:45 ` Victor Yodaiken
2001-08-26 21:00 ` Alan Cox
2001-08-26 20:42 ` Daniel Phillips
2001-08-26 19:31 ` Lehmann
2001-08-24 19:42 ` Lehmann
2001-08-24 21:42 ` Gérard Roudier
2001-08-25 0:05 ` Craig I. Hagan
-- strict thread matches above, loose matches on Subject: below --
2001-08-27 2:03 Rick Hohensee
2001-08-27 2:52 ` Keith Owens
2001-08-28 17:52 ` Kai Henningsen
2001-08-28 21:54 ` Matthew M
2001-08-28 1:08 Dieter Nützel
2001-08-28 0:05 ` Marcelo Tosatti
2001-08-28 1:54 ` Daniel Phillips
2001-08-28 5:01 ` Mike Galbraith
2001-08-28 18:18 ` Andrew Morton
2001-08-28 18:45 ` Hans Reiser
2001-08-28 15:28 Dieter Nützel
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=3B89F1FC.40F747D4@idb.hist.no \
--to=helgehaf@idb.hist.no \
--cc=linux-kernel@vger.kernel.org \
--cc=phillips@bonn-fries.net \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.