All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@osdl.org>
To: Andrea Arcangeli <andrea@suse.de>
Cc: nickpiggin@yahoo.com.au, axboe@suse.de, linux-kernel@vger.kernel.org
Subject: Re: Time sliced CFQ io scheduler
Date: Tue, 7 Dec 2004 18:00:33 -0800	[thread overview]
Message-ID: <20041207180033.6699425b.akpm@osdl.org> (raw)
In-Reply-To: <20041208013732.GF16322@dualathlon.random>

Andrea Arcangeli <andrea@suse.de> wrote:
>
> The desktop is ok with "as" simply because it's
>  normally optimal to stop writes completely

AS doesn't "stop writes completely".  With the current settings it
apportions about 1/3 of the disk's bandwidth to writes.

This thing Jens has found is for direct-io writes only.  It's a bug.

The other problem with AS is that it basically doesn't work at all with a
TCQ depth greater than four or so, and lots of people blindly look at
untuned SCSI benchmark results without realising that.  If a distro is
always selecting CFQ then they've probably gone and deoptimised all their
IDE users.  

AS needs another iteration of development to fix these things.  Right now
it's probably the case that we need CFQ or deadline for servers and AS for
desktops.   That's awkward.

  parent reply	other threads:[~2004-12-08  2:07 UTC|newest]

Thread overview: 66+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-12-02 13:04 Time sliced CFQ io scheduler Jens Axboe
2004-12-02 13:48 ` Jens Axboe
2004-12-02 19:48   ` Andrew Morton
2004-12-02 19:52     ` Jens Axboe
2004-12-02 20:19       ` Andrew Morton
2004-12-02 20:19         ` Jens Axboe
2004-12-02 20:34           ` Andrew Morton
2004-12-02 20:37             ` Jens Axboe
2004-12-07 23:11               ` Nick Piggin
2004-12-02 22:18         ` Prakash K. Cheemplavam
2004-12-03  7:01           ` Jens Axboe
2004-12-03  9:12             ` Prakash K. Cheemplavam
2004-12-03  9:18               ` Jens Axboe
2004-12-03  9:35                 ` Prakash K. Cheemplavam
2004-12-03  9:43                   ` Jens Axboe
2004-12-03  9:26               ` Andrew Morton
2004-12-03  9:34                 ` Prakash K. Cheemplavam
2004-12-03  9:39                 ` Jens Axboe
2004-12-03  9:54                   ` Prakash K. Cheemplavam
     [not found]                   ` <41B03722.5090001@gmx.de>
2004-12-03 10:31                     ` Jens Axboe
2004-12-03 10:38                       ` Jens Axboe
2004-12-03 10:45                         ` Prakash K. Cheemplavam
2004-12-03 10:48                           ` Jens Axboe
2004-12-03 11:27                             ` Prakash K. Cheemplavam
2004-12-03 11:29                               ` Jens Axboe
2004-12-03 11:52                                 ` Prakash K. Cheemplavam
2004-12-08  0:37       ` Andrea Arcangeli
2004-12-08  0:54         ` Nick Piggin
2004-12-08  1:37           ` Andrea Arcangeli
2004-12-08  1:47             ` Nick Piggin
2004-12-08  2:09               ` Andrea Arcangeli
2004-12-08  2:11                 ` Andrew Morton
2004-12-08  2:22                   ` Andrea Arcangeli
2004-12-08  6:52               ` Jens Axboe
2004-12-08  2:00             ` Andrew Morton [this message]
2004-12-08  2:08               ` Andrew Morton
2004-12-08  6:55                 ` Jens Axboe
2004-12-08  2:20               ` Andrea Arcangeli
2004-12-08  2:25                 ` Andrew Morton
2004-12-08  2:33                   ` Andrea Arcangeli
2004-12-08  2:33                   ` Nick Piggin
2004-12-08  2:51                     ` Andrea Arcangeli
2004-12-08  3:02                       ` Nick Piggin
2004-12-08  6:58                     ` Jens Axboe
2004-12-08  7:14                       ` Nick Piggin
2004-12-08  7:20                         ` Jens Axboe
2004-12-08  7:29                           ` Nick Piggin
2004-12-08  7:32                             ` Jens Axboe
2004-12-08  7:30                           ` Andrew Morton
2004-12-08  7:36                             ` Jens Axboe
2004-12-08 13:48                         ` Jens Axboe
2004-12-08  6:55               ` Jens Axboe
2004-12-08  7:08                 ` Nick Piggin
2004-12-08  7:11                   ` Jens Axboe
2004-12-08  7:19                     ` Nick Piggin
2004-12-08  7:26                       ` Jens Axboe
2004-12-08  9:35                         ` Jens Axboe
2004-12-08 10:08                           ` Jens Axboe
2004-12-08 12:47                           ` Jens Axboe
2004-12-08 10:52                 ` Helge Hafting
2004-12-08 10:49                   ` Jens Axboe
2004-12-08  6:49           ` Jens Axboe
2004-12-02 14:28 ` Giuliano Pochini
2004-12-02 14:41   ` Jens Axboe
2004-12-04 13:05     ` Giuliano Pochini
  -- strict thread matches above, loose matches on Subject: below --
2004-12-03 20:52 Chuck Ebbert

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=20041207180033.6699425b.akpm@osdl.org \
    --to=akpm@osdl.org \
    --cc=andrea@suse.de \
    --cc=axboe@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nickpiggin@yahoo.com.au \
    /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.