mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Andreas Dilger <adilger@sun.com>
To: Al Boldi <a1426z@gawab.com>
Cc: Jan Kara <jack@suse.cz>,
	linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
	Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
Subject: Re: [RFC] ext3: per-process soft-syncing data=ordered mode
Date: Wed, 06 Feb 2008 19:00:20 -0500	[thread overview]
Message-ID: <20080207000020.GB3149@webber.adilger.int> (raw)
In-Reply-To: <200801260827.59675.a1426z@gawab.com>

On Jan 26, 2008  08:27 +0300, Al Boldi wrote:
> Jan Kara wrote:
> > > data=ordered mode has proven reliable over the years, and it does this
> > > by ordering filedata flushes before metadata flushes.  But this
> > > sometimes causes contention in the order of a 10x slowdown for certain
> > > apps, either due to the misuse of fsync or due to inherent behaviour
> > > like db's, as well as inherent starvation issues exposed by the
> > > data=ordered mode.
> > >
> > > data=writeback mode alleviates data=order mode slowdowns, but only works
> > > per-mount and is too dangerous to run as a default mode.
> > >
> > > This RFC proposes to introduce a tunable which allows to disable fsync
> > > and changes ordered into writeback writeout on a per-process basis like
> > > this:
> > >
> > >       echo 1 > /proc/`pidof process`/softsync
> >
> >   I guess disabling fsync() was already commented on enough. Regarding
> > switching to writeback mode on per-process basis - not easily possible
> > because sometimes data is not written out by the process which stored
> > them (think of mmaped file).
> 
> Do you mean there is a locking problem?
> 
> > And in case of DB, they use direct-io
> > anyway most of the time so they don't care about journaling mode anyway.
> 
> Testing with sqlite3 and mysql4 shows that performance drastically improves 
> with writeback writeout.
> 
> >  But as Diego wrote, there is definitely some room for improvement in
> > current data=ordered mode so the difference shouldn't be as big in the
> > end.
> 
> Yes, it would be nice to get to the bottom of this starvation problem, but 
> even then, the proposed tunable remains useful for misbehaving apps.

Al, can you try a patch posted to linux-fsdevel and linux-ext4 from
Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> to see if this improves
your situation?  Dated Mon, 04 Feb 2008 19:15:25 +0900.

    [PATCH] ext3,4:fdatasync should skip metadata writeout when overwriting

It may be that we already have a solution in that patch for database
workloads where the pages are already allocated by avoiding the need
for ordered mode journal flushing in that case.

Cheers, Andreas
--
Andreas Dilger
Sr. Staff Engineer, Lustre Group
Sun Microsystems of Canada, Inc.


  parent reply	other threads:[~2008-02-07  0:33 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-01-24 20:36 Al Boldi
2008-01-24 21:50 ` Diego Calleja
2008-01-26  5:27   ` Al Boldi
2008-01-28 17:34     ` Jan Kara
2008-01-24 21:58 ` Valdis.Kletnieks
2008-01-26  5:27   ` Al Boldi
2008-01-25  1:19 ` Chris Snook
2008-01-26  5:28   ` Al Boldi
2008-01-29 17:22     ` Jan Kara
2008-01-30  6:04       ` Al Boldi
2008-01-30 14:29         ` Chris Mason
2008-01-30 18:39           ` Al Boldi
2008-01-31  0:32           ` Andreas Dilger
2008-01-31  6:20             ` Al Boldi
2008-01-31 16:56               ` Chris Mason
2008-01-31 17:10                 ` Jan Kara
2008-01-31 17:14                   ` Chris Mason
2008-02-01 21:26                     ` Al Boldi
2008-02-04 17:54                       ` Jan Kara
2008-02-05  7:07                         ` Al Boldi
2008-02-05 15:07                           ` Jan Kara
2008-02-05 19:20                             ` Al Boldi
2008-01-25  6:47 ` Andreas Dilger
2008-01-25 21:57   ` david
2008-01-25 15:36 ` Jan Kara
2008-01-26  5:27   ` Al Boldi
2008-01-28 17:27     ` Jan Kara
2008-01-28 20:17       ` Al Boldi
2008-02-07  0:00     ` Andreas Dilger [this message]
2008-02-10 14:54       ` Al Boldi

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=20080207000020.GB3149@webber.adilger.int \
    --to=adilger@sun.com \
    --cc=a1426z@gawab.com \
    --cc=hifumi.hisashi@oss.ntt.co.jp \
    --cc=jack@suse.cz \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.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

all inboxes | Powered by JetHome®