From: Gui Jianfeng <guijianfeng@cn.fujitsu.com>
To: Shaohua Li <shaohua.li@intel.com>
Cc: linux-kernel@vger.kernel.org, jens.axboe@oracle.com,
jmoyer@redhat.com, czoccolo@gmail.com, yanmin.zhang@intel.com
Subject: Re: cfq-iosched: tiobench regression
Date: Thu, 24 Dec 2009 15:48:38 +0800 [thread overview]
Message-ID: <4B331CD6.1070207@cn.fujitsu.com> (raw)
In-Reply-To: <20091224005506.GA7879@sli10-desk.sh.intel.com>
Shaohua Li wrote:
> We see about 30% regression in tiobench 32 threads 80M file sequential read.
> The regression is caused by below commits.
>
> 5db5d64277bf390056b1a87d0bb288c8b8553f96
> The commit makes the slice too small. In the test, the slice is limitted
> to 2 * idle_slice(300ms/32 < 2*idle_slice). This dramatically impacts io
> thoughput. The low_latency knob used to be only impact random io, now it
> impacts sequential io too. Any idea to fix it?
Hi Shaohua,
IMHO this shouldn't be a problem. Currently, low_latency is used to improve
the latency for the whole system. If someone would like to achieve high throughput,
just turn off low_latency knob.
next prev parent reply other threads:[~2009-12-24 7:53 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-12-24 0:55 cfq-iosched: tiobench regression Shaohua Li
2009-12-24 7:48 ` Gui Jianfeng [this message]
2009-12-24 9:19 ` Shaohua Li
2009-12-24 11:40 ` Corrado Zoccolo
2009-12-25 10:16 ` Corrado Zoccolo
2009-12-28 2:02 ` Shaohua Li
2009-12-28 2:03 ` [PATCH]cfq-iosched: don't take requests with long distence as close Shaohua Li
2009-12-28 8:36 ` Corrado Zoccolo
2009-12-28 8:46 ` Shaohua Li
2009-12-28 9:11 ` Corrado Zoccolo
2009-12-28 9:28 ` Shaohua Li
2009-12-28 9:40 ` Corrado Zoccolo
2009-12-28 12:16 ` Jens Axboe
2010-01-04 14:58 ` Jeff Moyer
2010-01-05 21:16 ` Jeff Moyer
2010-01-06 1:19 ` Li, Shaohua
2010-01-07 13:44 ` Corrado Zoccolo
2010-01-07 14:30 ` Jeff Moyer
2010-01-11 5:20 ` Zhang, Yanmin
2010-01-11 15:05 ` Corrado Zoccolo
2010-01-12 2:43 ` Zhang, Yanmin
2010-01-15 19:32 ` Corrado Zoccolo
2010-01-15 19:45 ` Jeff Moyer
2010-01-15 20:24 ` Corrado Zoccolo
2010-01-15 20:26 ` Jeff Moyer
2009-12-28 3:19 ` [PATCH]cfq-iosched: split seeky coop queues after one slice Shaohua Li
2009-12-28 8:40 ` Corrado Zoccolo
2009-12-28 8:52 ` Shaohua Li
2010-01-04 15:04 ` Jeff Moyer
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=4B331CD6.1070207@cn.fujitsu.com \
--to=guijianfeng@cn.fujitsu.com \
--cc=czoccolo@gmail.com \
--cc=jens.axboe@oracle.com \
--cc=jmoyer@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=shaohua.li@intel.com \
--cc=yanmin.zhang@intel.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.