From: Jeff Garzik <jeff@garzik.org>
To: Theodore Tso <tytso@mit.edu>, Jan Kara <jack@suse.cz>,
Chris Mason <chris.mason@oracle.com>,
Ric Wheeler <rwheeler@redhat.com>,
Linux Kernel Developers List <linux-kernel@vger.kernel.
Subject: Re: [PATCH 0/3] Ext3 latency improvement patches
Date: Fri, 27 Mar 2009 20:14:04 -0400 [thread overview]
Message-ID: <49CD6BCC.6080602@garzik.org> (raw)
In-Reply-To: <20090327230902.GG5176@mit.edu>
Theodore Tso wrote:
> OTOH, the really big databases will tend to use direct I/O, so they
> won't be dirtying the page cache anyway. So maybe it's not worth the
Not necessarily... From what I understand, a lot of the individual
low-level components in cloud storage, such as GoogleFS's chunk
server[1] do not bypass the page cache, even though they do care about
the details of data caching and data consistency.
I am looking at the same areas for my own distributed storage work, and
am finding that the current crop of Linux-specific,
database/server-friendly syscalls permit more application control over
pagecache usage than in past years, decreasing the need for O_DIRECT.
Things like readahead(2), sync_file_range(2), fadvise(3), really help.
Jeff
[1] http://labs.google.com/papers/gfs-sosp2003.pdf
WARNING: multiple messages have this Message-ID (diff)
From: Jeff Garzik <jeff@garzik.org>
To: Theodore Tso <tytso@mit.edu>, Jan Kara <jack@suse.cz>,
Chris Mason <chris.mason@oracle.com>,
Ric Wheeler <rwheeler@redhat.com>,
Linux Kernel Developers List <linux-kernel@vger.kernel.org>,
Ext4 Developers List <linux-ext4@vger.kernel.org>
Subject: Re: [PATCH 0/3] Ext3 latency improvement patches
Date: Fri, 27 Mar 2009 20:14:04 -0400 [thread overview]
Message-ID: <49CD6BCC.6080602@garzik.org> (raw)
In-Reply-To: <20090327230902.GG5176@mit.edu>
Theodore Tso wrote:
> OTOH, the really big databases will tend to use direct I/O, so they
> won't be dirtying the page cache anyway. So maybe it's not worth the
Not necessarily... From what I understand, a lot of the individual
low-level components in cloud storage, such as GoogleFS's chunk
server[1] do not bypass the page cache, even though they do care about
the details of data caching and data consistency.
I am looking at the same areas for my own distributed storage work, and
am finding that the current crop of Linux-specific,
database/server-friendly syscalls permit more application control over
pagecache usage than in past years, decreasing the need for O_DIRECT.
Things like readahead(2), sync_file_range(2), fadvise(3), really help.
Jeff
[1] http://labs.google.com/papers/gfs-sosp2003.pdf
next prev parent reply other threads:[~2009-03-28 0:14 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-03-27 20:24 [PATCH 0/3] Ext3 latency improvement patches Theodore Ts'o
2009-03-27 20:24 ` [PATCH 1/3] block_write_full_page: Use synchronous writes for WBC_SYNC_ALL writebacks Theodore Ts'o
2009-03-27 20:24 ` [PATCH 2/3] ext3: Use WRITE_SYNC for commits which are caused by fsync() Theodore Ts'o
2009-03-27 20:24 ` [PATCH 3/3] ext3: Avoid starting a transaction in writepage when not necessary Theodore Ts'o
2009-03-27 22:23 ` Jan Kara
2009-03-27 23:03 ` Theodore Tso
2009-03-30 13:22 ` Jan Kara
2009-03-30 13:22 ` Jan Kara
2009-03-27 22:20 ` [PATCH 2/3] ext3: Use WRITE_SYNC for commits which are caused by fsync() Jan Kara
2009-03-27 20:55 ` [PATCH 1/3] block_write_full_page: Use synchronous writes for WBC_SYNC_ALL writebacks Jan Kara
2009-04-07 6:21 ` Andrew Morton
2009-04-07 6:50 ` Andrew Morton
2009-04-07 6:50 ` Andrew Morton
2009-04-07 7:08 ` Jens Axboe
2009-04-07 7:17 ` Jens Axboe
2009-04-07 8:16 ` Jens Axboe
2009-04-07 7:23 ` Andrew Morton
2009-04-07 7:57 ` Jens Axboe
2009-04-07 19:09 ` Theodore Tso
2009-04-07 19:32 ` Jens Axboe
2009-04-07 21:44 ` Theodore Tso
2009-04-07 22:19 ` [PATCH] block_write_full_page: switch synchronous writes to use WRITE_SYNC_PLUG Theodore Tso
2009-04-07 22:19 ` Theodore Tso
2009-04-07 23:09 ` Andrew Morton
2009-04-07 23:46 ` Theodore Tso
2009-04-08 8:08 ` Jens Axboe
2009-04-08 22:34 ` Andrew Morton
2009-04-09 17:59 ` Jens Axboe
2009-04-08 6:00 ` Jens Axboe
2009-04-08 15:26 ` Theodore Tso
2009-04-08 5:58 ` [PATCH 1/3] block_write_full_page: Use synchronous writes for WBC_SYNC_ALL writebacks Jens Axboe
2009-04-08 15:25 ` Theodore Tso
2009-04-07 14:19 ` Theodore Tso
2009-03-27 20:50 ` [PATCH 0/3] Ext3 latency improvement patches Chris Mason
2009-03-27 21:03 ` Chris Mason
2009-03-27 21:19 ` Jan Kara
2009-03-27 21:30 ` Theodore Tso
2009-03-27 21:54 ` Jan Kara
2009-03-27 21:54 ` Jan Kara
2009-03-27 23:09 ` Theodore Tso
2009-03-28 0:14 ` Jeff Garzik [this message]
2009-03-28 0:14 ` Jeff Garzik
2009-03-28 0:24 ` David Rees
2009-03-28 0:24 ` David Rees
2009-03-30 14:16 ` Ric Wheeler
2009-03-30 11:23 ` Aneesh Kumar K.V
2009-03-30 11:23 ` Aneesh Kumar K.V
2009-03-30 11:44 ` Chris Mason
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=49CD6BCC.6080602@garzik.org \
--to=jeff@garzik.org \
--cc=chris.mason@oracle.com \
--cc=jack@suse.cz \
--cc=linux-kernel@vger.kernel. \
--cc=rwheeler@redhat.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.