From: Mark Lord <kernel@teksavvy.com>
To: Justin Piszcz <jpiszcz@lucidpixels.com>
Cc: Christoph Hellwig <hch@infradead.org>,
Alex Elder <aelder@sgi.com>,
Linux Kernel <linux-kernel@vger.kernel.org>,
Stan Hoeppner <stan@hardwarefreak.com>,
xfs@oss.sgi.com
Subject: Re: xfs: very slow after mount, very slow at umount
Date: Thu, 27 Jan 2011 11:03:49 -0500 [thread overview]
Message-ID: <4D419765.4070805@teksavvy.com> (raw)
In-Reply-To: <alpine.DEB.2.00.1101271040000.31246@p34.internal.lan>
On 11-01-27 10:40 AM, Justin Piszcz wrote:
>
>
> On Thu, 27 Jan 2011, Mark Lord wrote:
..
>> Can you recommend a good set of mkfs.xfs parameters to suit the characteristics
>> of this system? Eg. Only a few thousand active inodes, and nearly all files are
>> in the 600MB -> 20GB size range. The usage pattern it must handle is up to
>> six concurrent streaming writes at the same time as up to three streaming reads,
>> with no significant delays permitted on the reads.
>>
>> That's the kind of workload that I find XFS handles nicely,
>> and EXT4 has given me trouble with in the past.
..
> I did a load of benchmarks a long time ago testing every mkfs.xfs option there
> was, and I found that most of the time (if not all), the defaults were the best.
..
I am concerned with fragmentation on the very special workload in this case.
I'd really like the 20GB files, written over a 1-2 hour period, to consist
of a very few very large extents, as much as possible.
Rather than hundreds or thousands of "tiny" MB sized extents.
I wonder what the best mkfs.xfs parameters might be to encourage that?
Cheers
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
next prev parent reply other threads:[~2011-01-27 16:01 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <4D40C8D1.8090202@teksavvy.com>
2011-01-27 3:30 ` xfs: very slow after mount, very slow at umount Dave Chinner
2011-01-27 3:49 ` Mark Lord
2011-01-27 5:17 ` Stan Hoeppner
2011-01-27 15:12 ` Mark Lord
2011-01-27 15:40 ` Justin Piszcz
2011-01-27 16:03 ` Mark Lord [this message]
2011-01-27 19:40 ` Stan Hoeppner
2011-01-27 20:11 ` david
2011-01-27 23:53 ` Stan Hoeppner
2011-01-28 2:09 ` david
2011-01-28 13:56 ` Dave Chinner
2011-01-28 19:26 ` david
2011-01-29 5:40 ` Dave Chinner
2011-01-29 6:08 ` david
2011-01-29 7:35 ` Dave Chinner
2011-01-31 19:17 ` Christoph Hellwig
2011-01-27 21:56 ` Mark Lord
2011-01-28 0:17 ` Dave Chinner
2011-01-28 1:22 ` Mark Lord
2011-01-28 1:36 ` Mark Lord
2011-01-28 4:14 ` David Rees
2011-01-28 14:22 ` Mark Lord
2011-01-28 7:31 ` Dave Chinner
2011-01-28 14:33 ` Mark Lord
2011-01-28 23:58 ` Dave Chinner
2011-01-28 19:18 ` Martin Steigerwald
2011-01-27 20:24 ` John Stoffel
2011-01-27 23:41 ` Dave Chinner
2011-01-28 0:59 ` Mark Lord
2011-01-27 23:39 ` Dave Chinner
[not found] ` <4D40CDCF.4010301@teksavvy.com>
2011-01-27 3:43 ` Dave Chinner
2011-01-27 3:53 ` Mark Lord
2011-01-27 4:54 ` Mark Lord
2011-01-27 23:34 ` Dave Chinner
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=4D419765.4070805@teksavvy.com \
--to=kernel@teksavvy.com \
--cc=aelder@sgi.com \
--cc=hch@infradead.org \
--cc=jpiszcz@lucidpixels.com \
--cc=linux-kernel@vger.kernel.org \
--cc=stan@hardwarefreak.com \
--cc=xfs@oss.sgi.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