From: Ric Wheeler <rwheeler@redhat.com>
To: "Theodore Ts'o" <tytso@mit.edu>,
Chris Mason <chris.mason@fusionio.com>,
Chris Mason <clmason@fusionio.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Ingo Molnar <mingo@kernel.org>,
Christoph Hellwig <hch@infradead.org>,
Martin Steigerwald <Martin@lichtvoll.de>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Dave Chinner <david@fromorbit.com>,
linux-fsdevel <linux-fsdevel@vger.kernel.org>
Subject: Re: [PATCH, 3.7-rc7, RESEND] fs: revert commit bbdd6808 to fallocate UAPI
Date: Fri, 07 Dec 2012 17:02:32 -0500 [thread overview]
Message-ID: <50C26778.7000500@redhat.com> (raw)
In-Reply-To: <20121207215727.GF29435@thunk.org>
On 12/07/2012 04:57 PM, Theodore Ts'o wrote:
> On Fri, Dec 07, 2012 at 04:42:06PM -0500, Ric Wheeler wrote:
>> The other things that I think we should try would be to convert over
>> larger chunks as we discussed on the list back in the summer (just
>> because the user writes 4KB does not mean that we cannot flip over
>> 1MB and zero that).
> Writing a megabyte is not free. If you assume that your HDD has a
> sustained write throughput of 100-125 MB/s, writing a megabyte will
> take 8-10ms. It might be a win if you amortize it over a large number
> of writes, but it doesn't help your 99.9 percentile latency numbers.
> (99.9 percentile latency numbers matters because eventually you'll
> have a user request which hits multiple serial long latency
> operations, and then the delay looks **really** user visible.)
>
> - Ted
Writing 4KB at a time to a disk cost XX units of time.
Writing to the same sector (especially for a HDD), cost XX units + a small amount.
I suggest that we try it out.
For SSD's, much better to use specific HW offload commands if possible like
WRITE_SAME (zeroed) or UNMAP/TRIM to get that performance boost since no actual
data is moved...
ric
next prev parent reply other threads:[~2012-12-07 22:02 UTC|newest]
Thread overview: 69+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-19 23:04 [PATCH] fs: revert commit bbdd6808 to fallocate UAPI Dave Chinner
2012-11-20 16:36 ` Christoph Hellwig
2012-11-26 0:28 ` [PATCH, 3.7-rc7, RESEND] " Dave Chinner
2012-11-26 2:55 ` Theodore Ts'o
2012-11-26 6:14 ` Tao Ma
2012-11-26 9:12 ` Dave Chinner
2012-12-05 10:48 ` Martin Steigerwald
2012-12-05 15:45 ` Linus Torvalds
2012-12-05 16:18 ` Martin Steigerwald
2012-12-05 16:33 ` Theodore Ts'o
2012-12-05 17:24 ` Martin Steigerwald
2012-12-05 17:34 ` Theodore Ts'o
2012-12-05 17:55 ` Martin Steigerwald
2012-12-06 0:42 ` Dave Chinner
2012-12-06 9:24 ` Martin Steigerwald
2012-12-05 18:25 ` Linus Torvalds
2012-12-06 1:14 ` Dave Chinner
2012-12-06 3:03 ` Linus Torvalds
2012-12-06 9:37 ` Martin Steigerwald
2012-12-07 1:08 ` Ingo Molnar
2012-12-07 2:40 ` Dave Chinner
2012-12-07 10:24 ` Martin Steigerwald
2012-12-06 12:06 ` Christoph Hellwig
2012-12-06 16:50 ` Theodore Ts'o
2012-12-07 1:57 ` Dave Chinner
2012-12-06 12:05 ` Christoph Hellwig
2012-12-07 1:16 ` Ingo Molnar
2012-12-07 3:19 ` Dave Chinner
2012-12-07 17:36 ` Ric Wheeler
2012-12-07 18:18 ` Linus Torvalds
2012-12-07 19:03 ` Chris Mason
2012-12-07 20:43 ` Theodore Ts'o
2012-12-07 21:09 ` Chris Mason
2012-12-07 21:27 ` Theodore Ts'o
2012-12-07 21:43 ` Chris Mason
2012-12-07 21:49 ` Ric Wheeler
2012-12-07 21:57 ` Chris Mason
2012-12-07 22:51 ` Eric Sandeen
2012-12-07 22:52 ` Eric Sandeen
2012-12-07 21:42 ` Ric Wheeler
2012-12-07 21:57 ` Theodore Ts'o
2012-12-07 22:02 ` Ric Wheeler [this message]
2012-12-08 0:39 ` Dave Chinner
2012-12-08 2:52 ` Joel Becker
2012-12-08 4:04 ` Dave Chinner
2012-12-08 0:17 ` Dave Chinner
2012-12-08 1:39 ` Chris Mason
2012-12-10 16:02 ` Chris Mason
2012-12-10 17:37 ` Theodore Ts'o
2012-12-10 18:05 ` Steven Whitehouse
2012-12-10 18:13 ` Theodore Ts'o
2012-12-10 18:20 ` Theodore Ts'o
2012-12-11 12:16 ` Steven Whitehouse
2012-12-11 22:09 ` Dave Chinner
2012-12-10 18:52 ` Ric Wheeler
2012-12-11 0:52 ` Dave Chinner
2012-12-07 19:30 ` Steven Rostedt
2012-12-07 21:14 ` Theodore Ts'o
2012-12-07 21:47 ` Ric Wheeler
2012-12-07 23:25 ` Howard Chu
2012-12-08 0:50 ` Dave Chinner
2012-12-08 13:52 ` Howard Chu
2012-12-08 14:02 ` Ric Wheeler
2012-12-07 22:01 ` Eric Sandeen
2012-12-09 21:37 ` Ric Wheeler
2012-11-26 11:53 ` Alan Cox
2012-11-26 14:43 ` Theodore Ts'o
2012-11-26 21:12 ` Dave Chinner
2012-11-27 13:44 ` Martin Steigerwald
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=50C26778.7000500@redhat.com \
--to=rwheeler@redhat.com \
--cc=Martin@lichtvoll.de \
--cc=chris.mason@fusionio.com \
--cc=clmason@fusionio.com \
--cc=david@fromorbit.com \
--cc=hch@infradead.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=torvalds@linux-foundation.org \
--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;
as well as URLs for NNTP newsgroup(s).