From: Chris Mason <chris.mason@oracle.com>
To: konishi.ryusuke@lab.ntt.co.jp
Cc: pavel@suse.cz, akpm@linux-foundation.org,
linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 01/27] nilfs2: add document
Date: Tue, 16 Sep 2008 09:38:18 -0400 [thread overview]
Message-ID: <20080916133818.GA7472@think.oraclecorp.com> (raw)
In-Reply-To: <20080916.051041.110880811.ryusuke@osrg.net>
On Tue, Sep 16, 2008 at 05:10:41AM +0900, konishi.ryusuke@lab.ntt.co.jp wrote:
> Hi!
>
> On Mon, 15 Sep 2008 11:54:27 +0200, Pavel Machek wrote:
> > Hi!
> >
> > > +NILFS2 is a log-structured file system (LFS) supporting continuous
> > > +snapshotting. In addition to versioning capability of the entire file
> > > +system, users can even restore files mistakenly overwritten or
> >
> > Hmm, undelete done right. Just one question... how slow/fast is it
> > compared to conventional filesystems (ext3?)?
> > Pavel
>
> After my first submission, Szabolcs Szakacsits showed benchmark
> results using compilebench.
>
> On Thu, 21 Aug 2008 00:25:55 +0300 (MET DST)
> Szabolcs Szakacsits <szaka@ntfs-3g.org> wrote:
> > I ran compilebench on kernel 2.6.26 with freshly formatted volumes.
> > The behavior of NILFS2 was interesting.
> >
> > Its peformance rapidly degrades to the lowest ever measured level
> > (< 1 MB/s) but after a while it recovers and gives consistent numbers.
> > However it's still very far from the current unstable btrfs performance.
> > The results are reproducible.
> >
> > MB/s Runtime (s)
> > ----- -----------
> > btrfs unstable 17.09 572
> > ext3 13.24 877
> > btrfs 0.16 12.33 793
> > nilfs2 2nd+ runs 11.29 674
> > ntfs-3g 8.55 865
> > reiserfs 8.38 966
> > nilfs2 1st run 4.95 3800
> > xfs 1.88 3901
>
> Accordint to his measurement, NILFS2 showed a very low performance
> on the first measument, and it recovered after a while.
>
> I still don't know the reason why NILFS2 shows such behaviour, and
> I'm thinking to follow the benchmark.
Normally, compilebench has read phases that time how quickly the FS can
read the files after a bunch of operations. The runs above didn't
include the read phase, but in order to be fair to all the filesystems,
compilebench figures out the native readdir order of the FS so it can
create files in the optimal order for each fs.
It does this by creating all the files in its datasets and using readdir
to find out what order the FS returns. The files are all deleted
and the real runs start.
It is possible that bad perf in the first compilebench run is from
cleanup or transaction commits being done after the deletions.
-chris
next prev parent reply other threads:[~2008-09-16 13:38 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-09-14 19:07 [PATCH 0/27] nilfs2: continuous snapshotting file system Ryusuke Konishi
2008-09-14 19:07 ` [PATCH 01/27] nilfs2: add document Ryusuke Konishi
2008-09-14 19:07 ` [PATCH 02/27] nilfs2: disk format and userland interface Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 03/27] nilfs2: add inode and other major structures Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 04/27] nilfs2: integrated block mapping Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 05/27] nilfs2: B-tree based " Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 06/27] nilfs2: direct " Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 07/27] nilfs2: B-tree node cache Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 08/27] nilfs2: buffer and page operations Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 09/27] nilfs2: meta data file Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 10/27] nilfs2: persistent object allocator Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 11/27] nilfs2: disk address translator Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 12/27] nilfs2: inode map file Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 13/27] nilfs2: checkpoint file Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 14/27] nilfs2: segment usage file Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 15/27] nilfs2: inode operations Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 16/27] nilfs2: file operations Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 17/27] nilfs2: directory entry operations Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 18/27] nilfs2: pathname operations Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 19/27] nilfs2: operations for the_nilfs core object Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 20/27] nilfs2: super block operations Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 21/27] nilfs2: segment buffer Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 22/27] nilfs2: segment constructor Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 23/27] nilfs2: recovery functions Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 24/27] nilfs2: another dat for garbage collection Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 25/27] nilfs2: block cache " Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 26/27] nilfs2: ioctl operations Ryusuke Konishi
2008-09-14 19:08 ` [PATCH 27/27] nilfs2: update makefile and Kconfig Ryusuke Konishi
2008-09-17 14:41 ` [PATCH 25/27] nilfs2: block cache for garbage collection Jörn Engel
2008-09-17 19:09 ` Ryusuke Konishi
2008-09-17 22:49 ` Jörn Engel
2008-09-20 10:43 ` Ryusuke Konishi
2008-09-20 11:04 ` Jörn Engel
2008-09-15 18:20 ` [PATCH 18/27] nilfs2: pathname operations Pekka Enberg
2008-09-16 5:31 ` konishi.ryusuke
2008-09-17 14:31 ` [PATCH 02/27] nilfs2: disk format and userland interface Jörn Engel
2008-09-17 15:51 ` Ryusuke Konishi
2008-09-15 9:54 ` [PATCH 01/27] nilfs2: add document Pavel Machek
2008-09-15 20:10 ` konishi.ryusuke
2008-09-16 13:38 ` Chris Mason [this message]
2008-09-17 14:54 ` Jörn Engel
2008-09-17 17:52 ` Ryusuke Konishi
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=20080916133818.GA7472@think.oraclecorp.com \
--to=chris.mason@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=konishi.ryusuke@lab.ntt.co.jp \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pavel@suse.cz \
/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).