All of lore.kernel.org
 help / color / mirror / Atom feed
From: Phillip Susi <psusi@ubuntu.com>
To: Jan Kara <jack@suse.cz>, Eric Sandeen <sandeen@redhat.com>,
	Theodore Ts'o <tytso@mit.edu>, Fredrick <fjohnber@zoho.com>,
	Ric Wheeler <rwheeler@redhat.com>,
	linux-ext4@vger.kernel.org, Andreas Dilger <adilger@dilger.ca>,
	wenqing.lz@taobao.com
Subject: Re: ext4_fallocate
Date: Tue, 03 Jul 2012 23:06:58 -0400	[thread overview]
Message-ID: <4FF3B352.3040707@ubuntu.com> (raw)
In-Reply-To: <20120704023634.GB16947@gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 07/03/2012 10:36 PM, Zheng Liu wrote:
> Actually the workload needs to flush the data after writting a small
> random bits.  This workload is met in our product system at Taobao.
> Thus, the application has to wait this write to be done.  Certainly if
> we don't flush the data, the problem won't happen but there is a risk
> that we could loss our data.

Ohh, I see now... you want lots of small, random, synchronous writes.  Then I think the only way to avoid the metadata overhead is with the unsafe stale data patch.  More importantly, this workload is going to have terrible performance no matter what the fs does, because even with all of the blocks initialized, you're still doing lots of seeking and not allowing the IO elevator to help.  Maybe the application can be redesigned so that it does not generate such pathological IO?

Or is this another case of userspace really needing access to barriers rather than using the big hammer of fsync?

Is there any technical reason why a barrier flag can't be added to the aio interface?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJP87NSAAoJEJrBOlT6nu75F7wH/0NrhBZyp+KwYuJpqM+hUs8o
1CSRXpw/OBOljH61kAT1zqRofMNA/6gF5EYnAB3usUiwjJk4eQu4kFCVastZrLRQ
Zm3H2yuSXbeoRvuSnNBbpjCPFQYIKWHcKjt2+pov853Kb2auYSyN2T6I7qUmLu7U
tMJtJekMi4HPz3snBrLGsvYRyy7rMLWwf3wd+G2SBrVNS/tKuRSRaw0FwXqDaoN9
Dc4FqWmVeicG9qQbszkal84IZo1YwSWMzeRvqjn+LER1XTJJaSzvttppBKKFsS2M
FVeMwTtRHJ3wl3jWYTyN/AwDUg3Pr/PX9/xmdn8DSGQG3eo7sBlqyPBzN6QYz5w=
=C2YA
-----END PGP SIGNATURE-----

  reply	other threads:[~2012-07-04  3:07 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-06-25  6:42 ext4_fallocate Fredrick
2012-06-25  7:33 ` ext4_fallocate Andreas Dilger
2012-06-28 15:12   ` ext4_fallocate Phillip Susi
2012-06-28 15:23     ` ext4_fallocate Eric Sandeen
2012-06-25  8:51 ` ext4_fallocate Zheng Liu
2012-06-25 19:04   ` ext4_fallocate Fredrick
2012-06-25 19:17   ` ext4_fallocate Theodore Ts'o
2012-06-26  1:23     ` ext4_fallocate Fredrick
2012-06-26 13:13     ` ext4_fallocate Ric Wheeler
2012-06-26 17:30       ` ext4_fallocate Theodore Ts'o
2012-06-26 18:06         ` ext4_fallocate Fredrick
2012-06-26 18:21         ` ext4_fallocate Ric Wheeler
2012-06-26 18:57           ` ext4_fallocate Ted Ts'o
2012-06-26 19:22             ` ext4_fallocate Ric Wheeler
2012-06-26 18:05       ` ext4_fallocate Fredrick
2012-06-26 18:59         ` ext4_fallocate Ted Ts'o
2012-06-26 19:30         ` ext4_fallocate Ric Wheeler
2012-06-26 19:57           ` ext4_fallocate Eric Sandeen
2012-06-26 20:44             ` ext4_fallocate Eric Sandeen
2012-06-27 15:14               ` ext4_fallocate Eric Sandeen
2012-06-27 19:30               ` ext4_fallocate Theodore Ts'o
2012-06-27 23:02                 ` ext4_fallocate Eric Sandeen
2012-06-28 11:27                   ` ext4_fallocate Ric Wheeler
2012-06-29 19:02                     ` ext4_fallocate Andreas Dilger
2012-07-02  3:03                       ` ext4_fallocate Zheng Liu
2012-06-28 12:48                   ` ext4_fallocate Theodore Ts'o
2012-07-02  3:16                   ` ext4_fallocate Zheng Liu
2012-07-02 16:33                     ` ext4_fallocate Eric Sandeen
2012-07-02 17:44                       ` ext4_fallocate Jan Kara
2012-07-02 17:48                         ` ext4_fallocate Ric Wheeler
2012-07-03 17:41                           ` ext4_fallocate Zheng Liu
2012-07-03 17:57                             ` ext4_fallocate Zach Brown
2012-07-04  2:23                               ` ext4_fallocate Zheng Liu
2012-07-02 18:01                         ` ext4_fallocate Theodore Ts'o
2012-07-03  9:30                           ` ext4_fallocate Jan Kara
2012-07-04  1:15                         ` ext4_fallocate Phillip Susi
2012-07-04  2:36                           ` ext4_fallocate Zheng Liu
2012-07-04  3:06                             ` Phillip Susi [this message]
2012-07-04  3:48                               ` ext4_fallocate Zheng Liu
2012-07-04 12:20                               ` ext4_fallocate Ric Wheeler
2012-07-04 13:25                                 ` ext4_fallocate Zheng Liu
2012-06-26 13:06 ` ext4_fallocate Eric Sandeen

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=4FF3B352.3040707@ubuntu.com \
    --to=psusi@ubuntu.com \
    --cc=adilger@dilger.ca \
    --cc=fjohnber@zoho.com \
    --cc=jack@suse.cz \
    --cc=linux-ext4@vger.kernel.org \
    --cc=rwheeler@redhat.com \
    --cc=sandeen@redhat.com \
    --cc=tytso@mit.edu \
    --cc=wenqing.lz@taobao.com \
    /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.