From: Andi Kleen <andi@firstfloor.org>
To: Ben Gamari <bgamari.foss@gmail.com>
Cc: Avi Kivity <avi@redhat.com>,
linux-btrfs@vger.kernel.org, Andi Kleen <andi@firstfloor.org>,
Arjan van de Ven <arjan@infradead.org>,
linux-kernel@vger.kernel.org, tytso@mit.edu, npiggin@suse.de,
mingo@elte.hu, Ruald Andreae <ruald.a@gmail.com>,
Jens Axboe <jens.axboe@oracle.com>, Olly Betts <olly@survex.com>,
martin f krafft <madduck@madduck.net>
Subject: Re: Poor interactive performance with I/O loads with fsync()ing
Date: Sun, 11 Apr 2010 19:20:07 +0200 [thread overview]
Message-ID: <20100411172007.GA11514@basil.fritz.box> (raw)
In-Reply-To: <4bc1fa59.c5c2f10a.776d.ffffcd91@mx.google.com>
> Has the reason for this been identified? Judging from the nature of metadata
> loads, it would seem that it should be substantially easier to implement
> fsync() efficiently.
By design a copy on write tree fs would need to flush a whole
tree hierarchy on a sync. btrfs avoids this by using a special
log for fsync, but that causes more overhead if you have that
log on the same disk. So IO subsystem will do more work.
It's a bit like JBD data journaling.
However it should not have the stalls inherent in ext3's journaling.
-Andi
--
ak@linux.intel.com -- Speaking for myself only.
next prev parent reply other threads:[~2010-04-11 17:20 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-16 15:31 Poor interactive performance with I/O loads with fsync()ing Ben Gamari
2010-03-17 1:24 ` tytso
2010-03-17 3:18 ` Ben Gamari
2010-03-17 3:30 ` tytso
2010-03-17 4:31 ` Ben Gamari
2010-03-26 3:16 ` Ben Gamari
2010-03-17 4:53 ` Nick Piggin
2010-03-17 9:37 ` Ingo Molnar
2010-03-26 3:31 ` Ben Gamari
2010-04-09 15:21 ` Ben Gamari
2010-03-26 3:28 ` Ben Gamari
2010-03-23 19:51 ` Jesper Krogh
2010-03-26 3:13 ` Ben Gamari
2010-03-28 1:20 ` Ben Gamari
2010-03-28 1:29 ` Ben Gamari
2010-03-28 3:42 ` Arjan van de Ven
2010-03-28 14:06 ` Ben Gamari
2010-03-28 22:08 ` Andi Kleen
2010-04-09 14:56 ` Ben Gamari
2010-04-11 15:03 ` Avi Kivity
2010-04-11 16:35 ` Ben Gamari
2010-04-11 17:20 ` Andi Kleen [this message]
2010-04-11 18:16 ` Thomas Gleixner
2010-04-11 18:42 ` Andi Kleen
2010-04-11 21:54 ` Thomas Gleixner
2010-04-11 23:43 ` Hans-Peter Jansen
2010-04-12 0:22 ` Dave Chinner
2010-04-14 18:40 ` Ric Wheeler
-- strict thread matches above, loose matches on Subject: below --
2010-03-23 11:28 Pawel S
2010-03-23 13:27 ` Jens Axboe
2010-03-26 3:35 ` Ben Gamari
2010-03-30 10:46 ` Pawel S
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=20100411172007.GA11514@basil.fritz.box \
--to=andi@firstfloor.org \
--cc=arjan@infradead.org \
--cc=avi@redhat.com \
--cc=bgamari.foss@gmail.com \
--cc=jens.axboe@oracle.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=madduck@madduck.net \
--cc=mingo@elte.hu \
--cc=npiggin@suse.de \
--cc=olly@survex.com \
--cc=ruald.a@gmail.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.