Linux Device Mapper development
 help / color / mirror / Atom feed
From: Mike Snitzer <snitzer@redhat.com>
To: Keith Busch <keith.busch@intel.com>
Cc: Jens Axboe <axboe@kernel.dk>,
	dm-devel@redhat.com, shivakrishna.merla@netapp.com,
	Bart Van Assche <bvanassche@acm.org>,
	jmoyer@redhat.com
Subject: Re: awful request merge results while simulating high IOPS multipath
Date: Wed, 25 Feb 2015 19:11:15 -0500	[thread overview]
Message-ID: <20150226001114.GA22484@redhat.com> (raw)
In-Reply-To: <alpine.LNX.2.00.1502252327150.32385@localhost.lm.intel.com>

On Wed, Feb 25 2015 at  6:57pm -0500,
Keith Busch <keith.busch@intel.com> wrote:

> On Wed, 25 Feb 2015, Mike Snitzer wrote:
> >On Wed, Feb 25 2015 at  1:17pm -0500,
> >Busch, Keith <keith.busch@intel.com> wrote:
> >>Our first bottleneck appears to be the device mapper's single lock
> >>request queue.
> >
> >Obviously if we switched dm-multipath over to blk-mq we'd eliminate
> >that.  I'll see how things go and will share any changes I come up
> >with.
> 
> Yes, I'm also looking at blk-mq. It appears conversion helps a lot.

Oh, so you've already started a conversion of request-based DM?

> I'm not sure though how many tags or h/w contexts to allocate to ensure
> there's enough but not too many. You can't use the underlying device's
> blk-tagset count (assuming you could even get to it) since those are
> potentially shared among many block devices.

I'm very new to the blk-mq model so I have some learning to do before
I'll be much help.

But there really won't be a one size fits all amount for those resources
will there?  A multipath device can have _a lot_ of underlying paths.

  reply	other threads:[~2015-02-26  0:11 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-02-24 16:44 [PATCH 0/4] dm: simplify request-based DM a bit and an RFC-like perf tweak Mike Snitzer
2015-02-24 16:44 ` [PATCH 1/4] dm: remove unnecessary wrapper around blk_lld_busy Mike Snitzer
2015-02-24 16:44 ` [PATCH 2/4] dm: remove request-based DM queue's lld_busy_fn hook Mike Snitzer
2015-02-24 16:44 ` [PATCH 3/4] dm: remove request-based logic from make_request_fn wrapper Mike Snitzer
2015-02-24 16:44 ` [RFC PATCH 4/4] dm: delay running the queue slightly during request completion Mike Snitzer
2015-02-24 16:51   ` Jens Axboe
2015-02-24 17:22     ` [RFC PATCH 4/4 v2] " Mike Snitzer
2015-02-24 17:52       ` Jens Axboe
2015-02-24 18:12         ` Mike Snitzer
2015-02-24 18:16           ` Jens Axboe
2015-02-24 18:32             ` Mike Snitzer
2015-02-25  0:56               ` awful request merge results while simulating high IOPS multipath Mike Snitzer
2015-02-25  4:14                 ` Keith Busch
2015-02-25 15:11                   ` Jens Axboe
2015-02-25 18:17                     ` Busch, Keith
2015-02-25 22:10                       ` Mike Snitzer
2015-02-25 23:57                         ` Keith Busch
2015-02-26  0:11                           ` Mike Snitzer [this message]
2015-02-26  0:28                             ` Keith Busch
2015-02-25  4:38                 ` FIXED! [was: awful request merge results while simulating high IOPS] multipath Mike Snitzer
2015-02-25  4:41                   ` Mike Snitzer

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=20150226001114.GA22484@redhat.com \
    --to=snitzer@redhat.com \
    --cc=axboe@kernel.dk \
    --cc=bvanassche@acm.org \
    --cc=dm-devel@redhat.com \
    --cc=jmoyer@redhat.com \
    --cc=keith.busch@intel.com \
    --cc=shivakrishna.merla@netapp.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox