From: jbarnes@sgi.com (Jesse Barnes)
To: "Chen, Kenneth W" <kenneth.w.chen@intel.com>
Cc: linux-kernel@vger.kernel.org, Andrew Morton <akpm@osdl.org>
Subject: Re: per-cpu blk_plug_list
Date: Tue, 2 Mar 2004 19:44:32 -0800 [thread overview]
Message-ID: <20040303034432.GA31277@sgi.com> (raw)
In-Reply-To: <B05667366EE6204181EABE9C1B1C0EB501F2AB4C@scsmsx401.sc.intel.com>
On Mon, Mar 01, 2004 at 01:18:40PM -0800, Chen, Kenneth W wrote:
> blk_plug_list/blk_plug_lock manages plug/unplug action. When you have
> lots of cpu simultaneously submits I/O, there are lots of movement with
> the device queue on and off that global list. Our measurement showed
> that blk_plug_lock contention prevents linux-2.6.3 kernel to scale pass
> beyond 40 thousand I/O per second in the I/O submit path.
This helped out our machines quite a bit too. Without the patch, we
weren't able to scale above 80000 IOPS, but now we exceed 110000 (and
parity with our internal XSCSI based tree).
Maybe the plug lists and locks should be per-device though, rather than
per-cpu? That would make the migration case easier I think. Is that
possible?
Thanks,
Jesse
next prev parent reply other threads:[~2004-03-03 3:44 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-03-01 21:18 per-cpu blk_plug_list Chen, Kenneth W
2004-03-01 22:06 ` Jens Axboe
2004-03-01 22:17 ` Andrew Morton
2004-03-01 22:31 ` Andrew Morton
[not found] ` <B05667366EE6204181EABE9C1B1C0EB501F2AB4C@scsmsx401.sc.intel.com.suse.lists.linux.kernel>
2004-03-01 22:36 ` Andi Kleen
2004-03-01 22:39 ` per-cpu blk_plug_list II Andi Kleen
2004-03-03 3:44 ` Jesse Barnes [this message]
2004-03-03 3:55 ` per-cpu blk_plug_list Andrew Morton
-- strict thread matches above, loose matches on Subject: below --
2004-03-03 4:20 Chen, Kenneth W
2004-03-03 4:20 ` Chen, Kenneth W
2004-03-03 5:13 ` Andrew Morton
2004-03-03 5:13 ` Andrew Morton
2004-03-03 9:45 ` Miquel van Smoorenburg
2004-03-03 9:54 ` Andrew Morton
2004-03-03 16:24 ` Miquel van Smoorenburg
2004-03-03 22:37 ` Miquel van Smoorenburg
2004-03-01 22:28 Chen, Kenneth W
2004-03-01 22:14 Chen, Kenneth W
[not found] <cistron.B05667366EE6204181EABE9C1B1C0EB50211E5C8@scsmsx401.sc.intel.com>
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=20040303034432.GA31277@sgi.com \
--to=jbarnes@sgi.com \
--cc=akpm@osdl.org \
--cc=kenneth.w.chen@intel.com \
--cc=linux-kernel@vger.kernel.org \
/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.