From: Song Liu <song@kernel.org>
To: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org,
linux-raid <linux-raid@vger.kernel.org>,
stable@vger.kernel.org, Larkin Lowrey <llowrey@nuclearwinter.com>,
Wilson Jonathan <i400sjon@gmail.com>,
Roger Heflin <rogerheflin@gmail.com>
Subject: Re: [PATCH] block: check more requests for multiple_queues in blk_attempt_plug_merge
Date: Thu, 10 Mar 2022 14:37:15 -0800 [thread overview]
Message-ID: <CAPhsuW5zX96VaBMu-o=JUqDz2KLRBcNFM_gEsT=tHjeYqrngSQ@mail.gmail.com> (raw)
In-Reply-To: <9516f407-bb91-093b-739d-c32bda1b5d8d@kernel.dk>
On Thu, Mar 10, 2022 at 2:15 PM Jens Axboe <axboe@kernel.dk> wrote:
>
> On 3/8/22 11:42 PM, Song Liu wrote:
> > RAID arrays check/repair operations benefit a lot from merging requests.
> > If we only check the previous entry for merge attempt, many merge will be
> > missed. As a result, significant regression is observed for RAID check
> > and repair.
> >
> > Fix this by checking more than just the previous entry when
> > plug->multiple_queues == true.
> >
> > This improves the check/repair speed of a 20-HDD raid6 from 19 MB/s to
> > 103 MB/s.
>
> Do the underlying disks not have an IO scheduler attached? Curious why
> the merges aren't being done there, would be trivial when the list is
> flushed out. Because if the perf difference is that big, then other
> workloads would be suffering they are that sensitive to being within a
> plug worth of IO.
The disks have mq-deadline by default. I also tried kyber, the result is the
same. Raid repair work sends IOs to all the HDDs in a round-robin manner.
If we only check the previous request, there isn't much opportunity for
merge. I guess other workloads may have different behavior?
> Between your two approaches, I do greatly prefer the first one though.
I also like the first one better. But I am not sure whether it will slow down
other workloads. We can probably also make the second one cleaner
with a new variation of blk_start_plug.
Thanks,
Song
next prev parent reply other threads:[~2022-03-10 22:37 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-03-09 6:42 [PATCH] block: check more requests for multiple_queues in blk_attempt_plug_merge Song Liu
2022-03-10 6:48 ` Christoph Hellwig
2022-03-10 7:23 ` Song Liu
2022-03-10 22:10 ` Song Liu
2022-03-10 22:12 ` Song Liu
2022-03-10 22:15 ` Jens Axboe
2022-03-10 22:37 ` Song Liu [this message]
2022-03-10 23:02 ` Jens Axboe
2022-03-10 23:33 ` Song Liu
2022-03-11 0:07 ` Jens Axboe
2022-03-11 0:31 ` Song Liu
2022-03-11 0:36 ` Jens Axboe
2022-03-11 0:38 ` Jens Axboe
2022-03-11 1:14 ` Ming Lei
2022-03-11 1:21 ` Jens Axboe
2022-03-11 1:32 ` Ming Lei
2022-03-11 1:35 ` Jens Axboe
2022-03-11 8:09 ` Wols Lists
2022-03-11 14:16 ` Jens Axboe
2022-03-11 16:59 ` Song Liu
2022-03-11 21:41 ` Paul Menzel
2022-03-11 22:40 ` Song Liu
2022-03-11 11:30 ` Wilson Jonathan
2022-03-11 15:58 ` Wols Lists
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='CAPhsuW5zX96VaBMu-o=JUqDz2KLRBcNFM_gEsT=tHjeYqrngSQ@mail.gmail.com' \
--to=song@kernel.org \
--cc=axboe@kernel.dk \
--cc=i400sjon@gmail.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-raid@vger.kernel.org \
--cc=llowrey@nuclearwinter.com \
--cc=rogerheflin@gmail.com \
--cc=stable@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;
as well as URLs for NNTP newsgroup(s).