public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Martin Sustrik <sustrik@fastmq.com>
To: Jeff Moyer <jmoyer@redhat.com>
Cc: Roger Heflin <rogerheflin@gmail.com>,
	Martin Lucina <mato@kotelna.sk>,
	linux-kernel@vger.kernel.org
Subject: Re: Higher than expected disk write(2) latency
Date: Wed, 02 Jul 2008 20:20:39 +0200	[thread overview]
Message-ID: <486BC6F7.2090709@fastmq.com> (raw)
In-Reply-To: <x49abh0p2y1.fsf@segfault.boston.devel.redhat.com>


> I thought you were doing I/O to the underlying block device.  If so,
> there's no need to open with O_SYNC.  You do, however, need to open the
> device with O_DIRECT and align your buffers (and buffer lengths)
> properly.

Yes, we are using O_DIRECT + aligning the buffers. However, usign O_SYNC 
vs. O_ASYNC seems to matter (different latency). Maybe this is because 
we are using librt?

> Which AIO interface are you using, libaio or librt?  How many I/Os are
> you queueing to the device?  You may want to take a look at aio-stress.c
> as a way to test your device (this uses libaio, the in-kernel AIO
> interface).

We are using librt, queueing 10240 writes, 512 bytes each.

We'll retry with libaio. Thanks for advice.

Martin

  reply	other threads:[~2008-07-02 18:20 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-06-28 12:11 Higher than expected disk write(2) latency Martin Lucina
2008-06-28 13:11 ` Roger Heflin
2008-06-30 18:10   ` Martin Sustrik
2008-06-30 19:02     ` Roger Heflin
2008-06-30 22:20       ` Martin Sustrik
2008-07-01  0:11         ` Bernd Eckenfels
2008-07-02 16:48       ` Martin Sustrik
2008-07-02 18:15         ` Jeff Moyer
2008-07-02 18:20           ` Martin Sustrik [this message]
2008-07-04  3:16             ` David Dillow
2008-07-02 21:33         ` Roger Heflin
2008-06-28 14:47 ` David Newall
2008-06-29 11:34   ` Martin Sustrik
2008-07-10  5:27 ` Andrew Morton
2008-07-10  8:12   ` Martin Sustrik
2008-07-10  8:14     ` Andrew Morton
2008-07-10 13:29       ` Chris Mason
2008-07-10 13:41         ` Martin Lucina
2008-07-10 14:01           ` Arjan van de Ven
2008-07-10 14:18             ` Chris Mason
2008-07-10  8:31     ` Alan Cox
2008-07-10 13:17       ` Martin Sustrik
2008-07-10 13:18         ` Andrew Morton
2008-07-11 15:17       ` Martin Sustrik
     [not found] <fa.OZMA74BZPX46rhnjz1am4hB786M@ifi.uio.no>
2008-06-30  6:41 ` Robert Hancock

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=486BC6F7.2090709@fastmq.com \
    --to=sustrik@fastmq.com \
    --cc=jmoyer@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mato@kotelna.sk \
    --cc=rogerheflin@gmail.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