From: Shaohua Li <shli@kernel.org>
To: Jens Axboe <axboe@fb.com>
Cc: Jeff Moyer <jmoyer@redhat.com>,
linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-block@vger.kernel.org
Subject: Re: [PATCH 6/6] writeback: throttle buffered writeback
Date: Tue, 22 Mar 2016 14:30:00 -0700 [thread overview]
Message-ID: <20160322213000.GA45596@kernel.org> (raw)
In-Reply-To: <56F1A8D0.4060403@fb.com>
On Tue, Mar 22, 2016 at 02:19:28PM -0600, Jens Axboe wrote:
> On 03/22/2016 02:12 PM, Jeff Moyer wrote:
> >Hi, Jens,
> >
> >Jens Axboe <axboe@fb.com> writes:
> >
> >>If the device has write back caching, 'wb_cache_delay' delays by
> >>this amount of usecs when a write completes before allowing more.
> >
> >What's the reason behind that?
>
> For classic write back caching, the cache can absorb a bunch of writes
> shortly, which means that the completion cost only shows a small part of the
> overall cost. This means that if we just throttle on completion, then when
> the device starts committing to media, then we'll end up starving other IO
> anyway. This knob is a way to attempt to tame that.
Does request size matter? I think it's yes. If request size will be accounted,
there will be issue how to evaluate IO cost of each request, which is hard.
Looks the throttling is done regardless if there is other IO running, which
could hurt writeback.
Thanks,
Shaohua
next prev parent reply other threads:[~2016-03-22 21:30 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-22 17:55 [PATCHSET][RFC] Make background writeback not suck Jens Axboe
2016-03-22 17:55 ` [PATCH 1/6] block: ensure we don't truncate top bits of the request command flags Jens Axboe
2016-03-22 18:59 ` Christoph Hellwig
2016-03-22 19:01 ` Jens Axboe
2016-03-25 2:08 ` Mike Christie
2016-03-25 4:18 ` Jens Axboe
2016-03-22 17:55 ` [PATCH 2/6] writeback: wb_start_writeback() should use WB_SYNC_ALL for WB_REASON_SYNC Jens Axboe
2016-03-22 21:34 ` Dave Chinner
2016-03-22 21:40 ` Jens Axboe
2016-03-22 21:51 ` Jens Axboe
2016-03-22 22:04 ` Dave Chinner
2016-03-22 22:07 ` Jens Axboe
2016-03-22 17:55 ` [PATCH 3/6] block: add ability to flag write back caching on a device Jens Axboe
2016-03-22 18:57 ` Christoph Hellwig
2016-03-22 18:59 ` Jens Axboe
2016-03-22 17:55 ` [PATCH 4/6] sd: inform block layer of write cache state Jens Axboe
2016-03-22 17:55 ` [PATCH 5/6] NVMe: " Jens Axboe
2016-03-22 17:55 ` [PATCH 6/6] writeback: throttle buffered writeback Jens Axboe
2016-03-22 20:12 ` Jeff Moyer
2016-03-22 20:19 ` Jens Axboe
2016-03-22 20:27 ` Jeff Moyer
2016-03-22 21:30 ` Shaohua Li [this message]
2016-03-22 21:35 ` Jens Axboe
2016-03-22 21:51 ` [PATCHSET][RFC] Make background writeback not suck Dave Chinner
2016-03-22 22:03 ` Jens Axboe
2016-03-22 22:31 ` Dave Chinner
2016-03-22 22:57 ` Jens Axboe
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=20160322213000.GA45596@kernel.org \
--to=shli@kernel.org \
--cc=axboe@fb.com \
--cc=jmoyer@redhat.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox