From: Chris Mason <chris.mason@oracle.com>
To: Jeff Moyer <jmoyer@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>, Jan Kara <jack@suse.cz>,
Boaz Harrosh <bharrosh@panasas.com>,
Mike Snitzer <snitzer@redhat.com>,
linux-scsi@vger.kernel.org, neilb@suse.de, dm-devel@redhat.com,
linux-fsdevel@vger.kernel.org, lsf-pc@lists.linux-foundation.org,
"Darrick J. Wong" <djwong@us.ibm.com>
Subject: Re: [Lsf-pc] [LSF/MM TOPIC] a few storage topics
Date: Tue, 24 Jan 2012 10:15:04 -0500 [thread overview]
Message-ID: <20120124151504.GQ4387@shiny> (raw)
In-Reply-To: <x49r4yq9suf.fsf@segfault.boston.devel.redhat.com>
On Mon, Jan 23, 2012 at 01:28:08PM -0500, Jeff Moyer wrote:
> Andrea Arcangeli <aarcange@redhat.com> writes:
>
> > On Mon, Jan 23, 2012 at 05:18:57PM +0100, Jan Kara wrote:
> >> requst granularity. Sure, big requests will take longer to complete but
> >> maximum request size is relatively low (512k by default) so writing maximum
> >> sized request isn't that much slower than writing 4k. So it works OK in
> >> practice.
> >
> > Totally unrelated to the writeback, but the merged big 512k requests
> > actually adds up some measurable I/O scheduler latencies and they in
> > turn slightly diminish the fairness that cfq could provide with
> > smaller max request size. Probably even more measurable with SSDs (but
> > then SSDs are even faster).
>
> Are you speaking from experience? If so, what workloads were negatively
> affected by merging, and how did you measure that?
https://lkml.org/lkml/2011/12/13/326
This patch is another example, although for a slight different reason.
I really have no idea yet what the right answer is in a generic sense,
but you don't need a 512K request to see higher latencies from merging.
-chris
next prev parent reply other threads:[~2012-01-24 16:51 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CABE8wws67dn0fwhTCs_XqH0g_CxGuT+hPQH9cVFe1xx5t_O9Jw@mail.gmail.com>
2012-01-17 20:06 ` [LSF/MM TOPIC] a few storage topics Mike Snitzer
2012-01-17 21:36 ` [Lsf-pc] " Jan Kara
2012-01-18 22:58 ` Darrick J. Wong
2012-01-18 23:22 ` Jan Kara
2012-01-18 23:42 ` Boaz Harrosh
2012-01-19 9:46 ` Jan Kara
2012-01-19 15:08 ` Andrea Arcangeli
2012-01-19 20:52 ` Jan Kara
2012-01-19 21:39 ` Andrea Arcangeli
2012-01-22 11:31 ` Boaz Harrosh
2012-01-23 16:30 ` Jan Kara
2012-01-22 12:21 ` Boaz Harrosh
2012-01-23 16:18 ` Jan Kara
2012-01-23 17:53 ` Andrea Arcangeli
2012-01-23 18:28 ` Jeff Moyer
2012-01-23 18:56 ` Andrea Arcangeli
2012-01-23 19:19 ` Jeff Moyer
2012-01-24 15:15 ` Chris Mason [this message]
2012-01-24 16:56 ` [dm-devel] " Christoph Hellwig
2012-01-24 17:01 ` Andreas Dilger
2012-01-24 17:06 ` [Lsf-pc] [dm-devel] " Andrea Arcangeli
2012-01-24 17:08 ` Chris Mason
2012-01-24 17:08 ` [Lsf-pc] " Andreas Dilger
2012-01-24 18:05 ` [dm-devel] " Jeff Moyer
2012-01-24 18:40 ` Christoph Hellwig
2012-01-24 19:07 ` Chris Mason
2012-01-24 19:14 ` Jeff Moyer
2012-01-24 20:09 ` [Lsf-pc] [dm-devel] " Jan Kara
2012-01-24 20:13 ` [Lsf-pc] " Jeff Moyer
2012-01-24 20:39 ` [Lsf-pc] [dm-devel] " Jan Kara
2012-01-24 20:59 ` Jeff Moyer
2012-01-24 21:08 ` Jan Kara
2012-01-25 3:29 ` Wu Fengguang
2012-01-25 6:15 ` [Lsf-pc] " Andreas Dilger
2012-01-25 6:35 ` [Lsf-pc] [dm-devel] " Wu Fengguang
2012-01-25 14:00 ` Jan Kara
2012-01-26 12:29 ` Andreas Dilger
2012-01-27 17:03 ` Ted Ts'o
2012-01-26 16:25 ` Vivek Goyal
2012-01-26 20:37 ` Jan Kara
2012-01-26 22:34 ` Dave Chinner
2012-01-27 3:27 ` Wu Fengguang
2012-01-27 5:25 ` Andreas Dilger
2012-01-27 7:53 ` Wu Fengguang
2012-01-25 14:33 ` Steven Whitehouse
2012-01-25 14:45 ` Jan Kara
2012-01-25 16:22 ` Loke, Chetan
2012-01-25 16:40 ` Steven Whitehouse
2012-01-25 17:08 ` Loke, Chetan
2012-01-25 17:32 ` James Bottomley
2012-01-25 18:28 ` Loke, Chetan
2012-01-25 18:37 ` Loke, Chetan
2012-01-25 18:37 ` James Bottomley
2012-01-25 20:06 ` Chris Mason
2012-01-25 22:46 ` Andrea Arcangeli
2012-01-25 22:58 ` Jan Kara
2012-01-26 8:59 ` Boaz Harrosh
2012-01-26 16:40 ` Loke, Chetan
2012-01-26 17:00 ` Andreas Dilger
2012-01-26 17:16 ` Loke, Chetan
2012-02-03 12:37 ` Wu Fengguang
2012-01-26 22:38 ` Dave Chinner
2012-01-26 16:17 ` Loke, Chetan
2012-01-25 18:44 ` Boaz Harrosh
2012-02-03 12:55 ` Wu Fengguang
2012-01-24 19:11 ` [dm-devel] [Lsf-pc] " Jeff Moyer
2012-01-26 22:31 ` Dave Chinner
2012-01-24 17:12 ` Jeff Moyer
2012-01-24 17:32 ` Chris Mason
2012-01-24 18:14 ` Jeff Moyer
2012-01-25 0:23 ` NeilBrown
2012-01-25 6:11 ` Andreas Dilger
2012-01-18 23:39 ` Dan Williams
2012-01-24 17:59 ` Martin K. Petersen
2012-01-24 19:48 ` Douglas Gilbert
2012-01-24 20:04 ` Martin K. Petersen
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=20120124151504.GQ4387@shiny \
--to=chris.mason@oracle.com \
--cc=aarcange@redhat.com \
--cc=bharrosh@panasas.com \
--cc=djwong@us.ibm.com \
--cc=dm-devel@redhat.com \
--cc=jack@suse.cz \
--cc=jmoyer@redhat.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=lsf-pc@lists.linux-foundation.org \
--cc=neilb@suse.de \
--cc=snitzer@redhat.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).