From: Jeremy Allison <jra@samba.org>
To: Theodore Tso <tytso@mit.edu>,
Andrew Morton <akpm@linux-foundation.org>,
"Cabot, Mason B" <mason.b.cabot@intel.com>,
linux-kernel@vger.kernel.org, tridge@samba.com, jra@samba.org
Subject: Re: Ext3 vs NTFS performance
Date: Wed, 2 May 2007 11:08:10 -0700 [thread overview]
Message-ID: <20070502180810.GC28726@jeremy-laptop> (raw)
In-Reply-To: <20070502161638.GB19442@thunk.org>
On Wed, May 02, 2007 at 12:16:38PM -0400, Theodore Tso wrote:
> On Tue, May 01, 2007 at 02:23:25PM -0700, Andrew Morton wrote:
> > On Tue, 1 May 2007 13:43:18 -0700
> > "Cabot, Mason B" <mason.b.cabot@intel.com> wrote:
> >
> > > I've been testing the NAS performance of ext3/Openfiler 2.2 against
> > > NTFS/WinXP and have found that NTFS significantly outperforms ext3 for
> > > video workloads. The Windows CIFS client will attempt a poor-man's
> > > pre-allocation of the file on the server by sending 1-byte writes at
> > > 128K-byte strides, breaking block allocation on ext3 and leading to
> > > fragmentation and poor performance. This will happen for many
> > > applications (including iTunes) as the CIFS client issues these
> > > pre-allocates under the application layer.
> >
> > Oh my gawd, what a stupid hack. Now we know what the MS interoperability
> > lab has been working on.
>
> I wonder if they patented this technique as well, as well as one of
> their dozen or so patents they are filing every day? "A Method of
> Screwing Over Samba's Performance So that Windows Longhorn Can Compete
> On Performance" coming soon, to a patent database near you! :-)
>
> > > I've posted a brief paper on Intel's OSS website
> > > (http://softwarecommunity.intel.com/articles/eng/1259.htm). Please give
> > > it a read and let me know what you think. In particular, I'd like to
> > > arrive at the right place to fix this problem: is it in the filesystem,
> > > VFS, or Samba?
>
> The right place is clearly Samba. I can't think of any other program
> or filesystem protocol where writing a 1 byte write at 128k strides
> would be used to signal a desire to do preallocation. In fact, it's
> hard to think of a worse way of doing things.
In fact they don't need to do this - there's an explicit CIFS
set file allocation call to pre-allocate size they could use.
There's a specific Samba VFS module that has XFS specific calls
to do this - vfs_prealloc. - but this won't work on ext3.
Jeremy.
next prev parent reply other threads:[~2007-05-02 18:08 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-05-01 20:43 Cabot, Mason B
2007-05-01 21:23 ` Andrew Morton
2007-05-02 12:21 ` Andi Kleen
2007-05-02 16:04 ` Theodore Tso
2007-05-02 18:40 ` Andi Kleen
2007-05-02 19:28 ` Theodore Tso
2007-05-02 16:16 ` Theodore Tso
2007-05-02 18:08 ` Jeremy Allison [this message]
2007-05-02 19:34 ` Theodore Tso
2007-05-02 20:38 ` Jeff Garzik
2007-05-02 22:01 ` Theodore Tso
2007-05-02 3:54 ` Gerhard Mack
2007-05-02 15:46 ` David Chinner
2007-05-02 15:44 ` David Chinner
2007-05-02 19:46 ` Chris Mason
2007-05-03 0:15 ` David Chinner
2007-05-03 12:57 ` Chris Mason
2007-05-03 21:14 ` Valerie Henson
2007-05-03 22:40 ` Bernd Eckenfels
2007-05-04 8:12 ` Anton Altaparmakov
2007-05-04 9:46 ` Christoph Hellwig
2007-05-04 14:47 ` Anton Altaparmakov
2007-05-04 15:49 ` Michael Tokarev
2007-05-04 18:41 ` Theodore Tso
2007-05-05 9:59 ` Christoph Hellwig
2007-05-06 20:59 ` Jörn Engel
2007-05-04 12:23 ` Theodore Tso
2007-05-04 19:40 ` Valerie Henson
2007-05-04 18:56 ` Phillip Susi
2007-05-04 19:52 ` Cabot, Mason B
2007-05-07 14:31 ` Phillip Susi
2007-09-12 23:47 ` Update: " Cabot, Mason B
2007-05-03 3:51 Al Boldi
2007-05-05 3:13 Xu CanHao
2007-05-05 13:45 ` Theodore Tso
[not found] <8hiYr-2fJ-1@gated-at.bofh.it>
[not found] ` <8huGm-2W4-33@gated-at.bofh.it>
2007-05-05 22:25 ` Bodo Eggert
2007-05-06 5:04 ` Xu CanHao
2007-05-06 1:48 Albert Cahalan
[not found] <8gShI-3hY-11@gated-at.bofh.it>
[not found] ` <8h1bh-8sG-11@gated-at.bofh.it>
[not found] ` <8h2Al-280-1@gated-at.bofh.it>
[not found] ` <8hW9y-2Lp-3@gated-at.bofh.it>
2007-05-07 11:21 ` Bodo Eggert
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=20070502180810.GC28726@jeremy-laptop \
--to=jra@samba.org \
--cc=akpm@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mason.b.cabot@intel.com \
--cc=tridge@samba.com \
--cc=tytso@mit.edu \
/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®