From: "Stephen C. Tweedie" <sct@redhat.com>
To: cmm@us.ibm.com
Cc: Badari Pulavarty <pbadari@us.ibm.com>,
Andreas Dilger <adilger@clusterfs.com>, Jan Kara <jack@suse.cz>,
"Theodore Ts'o" <tytso@mit.edu>, Andrew Morton <akpm@osdl.org>,
lkml <linux-kernel@vger.kernel.org>,
linux-fsdevel <linux-fsdevel@vger.kernel.org>,
Stephen Tweedie <sct@redhat.com>
Subject: Re: ext3_ordered_writepage() questions
Date: Mon, 20 Mar 2006 12:05:53 -0500 [thread overview]
Message-ID: <1142874353.3414.18.camel@orbit.scot.redhat.com> (raw)
In-Reply-To: <1142637789.4802.34.camel@dyn9047017067.beaverton.ibm.com>
Hi,
On Fri, 2006-03-17 at 15:23 -0800, Mingming Cao wrote:
> > > There is one other perspective to be aware of, though: the current
> > > behaviour means that by default ext3 generally starts flushing pending
> > > writeback data within 5 seconds of a write. Without that, we may end up
> > > accumulating a lot more dirty data in memory, shifting the task of write
> > > throttling from the filesystem to the VM.
> Current data=writeback mode already behaves like this, so the VM
> subsystem should be tested for a certain extent, isn't?
Yes, but there are repeated reports that for many workloads,
data=writeback is actually slower than data=ordered. So there are
probably some interactions like this which may be hurting us already.
--Stephen
next prev parent reply other threads:[~2006-03-20 17:07 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-03-08 0:19 [RFC PATCH 0/3] VFS changes to collapse all the vectored and AIO support Badari Pulavarty
2006-03-08 0:22 ` [PATCH 1/3] Vectorize aio_read/aio_write methods Badari Pulavarty
2006-03-08 12:44 ` christoph
2006-03-08 0:23 ` [PATCH 2/3] Remove readv/writev methods and use aio_read/aio_write instead Badari Pulavarty
2006-03-08 12:45 ` christoph
2006-03-08 16:26 ` Badari Pulavarty
2006-03-08 0:24 ` [PATCH 3/3] Zach's core aio changes to support vectored AIO Badari Pulavarty
2006-03-08 3:37 ` Benjamin LaHaise
2006-03-08 16:34 ` Badari Pulavarty
2006-03-08 12:47 ` [RFC PATCH 0/3] VFS changes to collapse all the vectored and AIO support christoph
2006-03-08 16:24 ` Badari Pulavarty
2006-03-09 16:17 ` ext3_ordered_writepage() questions Badari Pulavarty
2006-03-09 23:35 ` Andrew Morton
2006-03-10 0:36 ` Badari Pulavarty
2006-03-16 18:09 ` Theodore Ts'o
2006-03-16 18:22 ` Badari Pulavarty
2006-03-16 21:04 ` Theodore Ts'o
2006-03-16 21:57 ` Badari Pulavarty
2006-03-16 22:05 ` Jan Kara
2006-03-16 23:45 ` Badari Pulavarty
2006-03-17 0:44 ` Theodore Ts'o
2006-03-17 0:54 ` Andreas Dilger
2006-03-17 17:05 ` Stephen C. Tweedie
2006-03-17 21:32 ` Badari Pulavarty
2006-03-17 22:22 ` Stephen C. Tweedie
2006-03-17 22:38 ` Badari Pulavarty
2006-03-17 23:23 ` Mingming Cao
2006-03-20 17:05 ` Stephen C. Tweedie [this message]
2006-03-18 2:57 ` Suparna Bhattacharya
2006-03-18 3:02 ` Suparna Bhattacharya
2006-03-17 15:32 ` Jamie Lokier
2006-03-17 21:50 ` Stephen C. Tweedie
2006-03-17 22:11 ` Theodore Ts'o
2006-03-17 22:44 ` Jamie Lokier
2006-03-18 23:40 ` Theodore Ts'o
2006-03-19 2:36 ` Jamie Lokier
2006-03-19 5:28 ` Chris Adams
2006-03-20 2:18 ` Theodore Ts'o
2006-03-20 16:26 ` Stephen C. Tweedie
2006-03-17 22:23 ` Jamie Lokier
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=1142874353.3414.18.camel@orbit.scot.redhat.com \
--to=sct@redhat.com \
--cc=adilger@clusterfs.com \
--cc=akpm@osdl.org \
--cc=cmm@us.ibm.com \
--cc=jack@suse.cz \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbadari@us.ibm.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
Powered by JetHome