From: "Trenton D. Adams" <trenton.d.adams@gmail.com>
To: David Rees <drees76@gmail.com>
Cc: Christian Kujau <lists@nerdbynature.de>,
Artem Bityutskiy <Artem.Bityutskiy@nokia.com>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: EXT4-ish "fixes" in UBIFS
Date: Thu, 2 Apr 2009 20:26:01 -0600 [thread overview]
Message-ID: <9b1675090904021926o4a404c3ajf623b1560f919289@mail.gmail.com> (raw)
In-Reply-To: <72dbd3150904021855v440f46a7oc21a7ed28fbfcb13@mail.gmail.com>
On Thu, Apr 2, 2009 at 7:55 PM, David Rees <drees76@gmail.com> wrote:
> On Thu, Apr 2, 2009 at 5:28 PM, Trenton D. Adams
> <trenton.d.adams@gmail.com> wrote:
>> On Thu, Apr 2, 2009 at 6:24 PM, Trenton D. Adams
>> <trenton.d.adams@gmail.com> wrote:
>>> Yes, mounting "-o sync" does improve ext3 performance. It sucks
>>> though, because I do want quick writes. And mounting with sync option
>>> slows down to disk io speeds. In my case, that's between 20 and 23
>>> megabytes per second *big frown, quivering lip, and tears in my eyes*.
>>> :P
>>>
>>
>> Oh, I should have clarified. It improves performance under heavy
>> load. Under normal load, mounting without sync is fine. What I tend
>> to do is mount with "remount,rw,sync" when heavy load is starting.
>> Then my system goes slowly, but latency is good. Then, when it's all
>> done (say a big compile, or job, or whatever), I remount without sync
>> again.
>>
>> I'm thinking of writing a script that monitors performance, and
>> remounts as needed, lol. WHAT A HACK. hehe.
>
> All you're doing here is implementing the lowering of dirty data
> limits in the VM dynamically based on how long fsyncs take.
>
> Linus outlined this specific strategy as "the ideal siutation"
> somewhere in the depths of "That filesystem thread".
>
> Look at the new in 2.6.29 dirty*bytes parameters in
> Documentation/sysctl/vm.txt for more info. By lowering those values,
> you can effectively turn normal writes into synchronous writes which
> will greatly reduce latency of fsync under heavy write load.
WOW, that makes a huge difference. If I set it to 100M, I get the
10-15 second delay I was talking about. But, if I set it to 1M, I get
0.3 to 0.4 second delay on a 1M fsync. That is way better. Perhaps I
should auto-tune based on that parameter then. Although I do agree
with Linus that it sucks to do userland auto-tuning. :P
next prev parent reply other threads:[~2009-04-03 2:26 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-03-27 12:48 EXT4-ish "fixes" in UBIFS Artem Bityutskiy
2009-03-28 1:22 ` Kyungmin Park
2009-03-29 12:31 ` Artem Bityutskiy
2009-03-29 12:54 ` Artem Bityutskiy
2009-03-29 12:26 ` replace() system call needed (was Re: EXT4-ish "fixes" in UBIFS) Pavel Machek
2009-03-29 12:42 ` Artem Bityutskiy
2009-03-29 12:50 ` Pavel Machek
2009-03-29 13:00 ` Artem Bityutskiy
2009-03-29 13:02 ` Pavel Machek
2009-03-29 13:07 ` Artem Bityutskiy
2009-03-29 13:22 ` Andreas T.Auer
2009-03-29 13:55 ` Artem Bityutskiy
2009-03-29 13:40 ` Pavel Machek
2009-03-29 13:57 ` Artem Bityutskiy
2009-03-29 14:00 ` Pavel Machek
2009-03-30 17:19 ` Ric Wheeler
2009-03-30 22:11 ` Pavel Machek
2009-03-29 13:01 ` Andreas T.Auer
2009-03-29 13:06 ` Artem Bityutskiy
2009-03-30 15:58 ` Diego Calleja
2009-04-03 0:09 ` EXT4-ish "fixes" in UBIFS Christian Kujau
2009-04-03 0:24 ` Trenton D. Adams
2009-04-03 0:28 ` Trenton D. Adams
2009-04-03 0:38 ` Christian Kujau
2009-04-03 0:54 ` Trenton D. Adams
2009-04-03 0:54 ` Trenton D. Adams
2009-04-03 0:59 ` Trenton D. Adams
2009-04-03 1:55 ` David Rees
2009-04-03 2:05 ` Trenton D. Adams
2009-04-03 2:19 ` David Rees
2009-04-03 2:28 ` Trenton D. Adams
2009-04-03 2:58 ` David Rees
2009-04-03 3:13 ` Trenton D. Adams
2009-04-03 3:14 ` Trenton D. Adams
2009-04-03 5:02 ` Theodore Tso
2009-04-03 5:15 ` Trenton D. Adams
2009-04-03 6:30 ` Theodore Tso
2009-04-03 18:53 ` Chris Adams
2009-04-03 18:05 ` David Rees
2009-04-09 20:17 ` Pavel Machek
2009-04-03 2:26 ` Trenton D. Adams [this message]
2009-04-03 2:05 ` Theodore Tso
2009-04-03 2:45 ` Christian Kujau
2009-04-03 2:49 ` Trenton D. Adams
2009-04-03 6:53 ` Artem Bityutskiy
[not found] <cmFiD-8uc-9@gated-at.bofh.it>
[not found] ` <cmFss-ft-15@gated-at.bofh.it>
[not found] ` <cmFsu-ft-23@gated-at.bofh.it>
[not found] ` <cmGRt-2hq-7@gated-at.bofh.it>
[not found] ` <cmH1b-2K0-11@gated-at.bofh.it>
[not found] ` <cmHkz-3d3-5@gated-at.bofh.it>
[not found] ` <cmHkA-3d3-7@gated-at.bofh.it>
[not found] ` <cmHND-3Oz-5@gated-at.bofh.it>
[not found] ` <cmJPm-7hd-5@gated-at.bofh.it>
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=9b1675090904021926o4a404c3ajf623b1560f919289@mail.gmail.com \
--to=trenton.d.adams@gmail.com \
--cc=Artem.Bityutskiy@nokia.com \
--cc=drees76@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lists@nerdbynature.de \
/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).