Linux block layer
 help / color / mirror / Atom feed
From: Jens Axboe <axboe@kernel.dk>
To: Bart Van Assche <bvanassche@acm.org>
Cc: linux-block@vger.kernel.org, Josef Bacik <josef@toxicpanda.com>,
	Eric Dumazet <edumazet@kernel.org>,
	Aibing Chen <chenaibing@cmss.chinamobile.com>
Subject: Re: [PATCH 0/8] nbd: Defer request transmission to a workqueue
Date: Tue, 17 Nov 2026 16:09:49 -0700	[thread overview]
Message-ID: <cfbce18c-2a13-48ca-b2ac-938ad40ae892@kernel.dk> (raw)
In-Reply-To: <8de875b7-e7e7-47b8-8cf9-6d07c791a53c@acm.org>

On 11/17/26 4:03 PM, Bart Van Assche wrote:
> On 11/16/26 6:47 PM, Jens Axboe wrote:
>> Something like this _really_ needs performance numbers.
> 
> Hi Jens,
> 
> Measurements for the most extreme case (4 K IOPS; Unix socket) show about 5% less write IOPS (310 K -> 290 K). I think this is acceptable
> because anyone interested in peak performance will use drivers that
> yield higher performance, e.g. NVMe over TCP. I will publish the code
> that I used to gather these measurements soon.

OK, so I take that as -5% at high queue depth, what about qd=1 type of
results? Agree that nbd isn't the end all of networked performance, but
it's still nice to know what kind of performance hits we're talking
about. And either you didn't/don't know, or you're holding back on us a
bit here... Because I suspect the qd=1 results will look considerably
worse than the 5% drop.

-- 
Jens Axboe

  reply	other threads:[~2026-11-17 23:09 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-16 21:48 [PATCH 0/8] nbd: Defer request transmission to a workqueue Bart Van Assche
2026-11-16 21:48 ` [PATCH 1/8] nbd: Wake up config->conn_wait when setting NBD_RT_DISCONNECTED Bart Van Assche
2026-11-16 21:48 ` [PATCH 2/8] nbd: Requeue requests directly in nbd_send_cmd() Bart Van Assche
2026-11-16 21:48 ` [PATCH 3/8] nbd: Stop taking a config_refs reference per request Bart Van Assche
2026-11-16 21:48 ` [PATCH 4/8] nbd: Defer request transmission to a workqueue Bart Van Assche
2026-11-16 21:48 ` [PATCH 5/8] nbd: Call sk_set_memalloc() outside nsock->tx_lock Bart Van Assche
2026-11-16 21:48 ` [PATCH 6/8] nbd: Remove socket lockdep reclassification Bart Van Assche
2026-11-16 21:48 ` [PATCH 7/8] nbd: Remove the partial-send handling Bart Van Assche
2026-11-16 21:48 ` [PATCH 8/8] nbd: Remove the sent argument of sock_xmit() Bart Van Assche
2026-11-17  2:47 ` [PATCH 0/8] nbd: Defer request transmission to a workqueue Jens Axboe
2026-11-17 23:03   ` Bart Van Assche
2026-11-17 23:09     ` Jens Axboe [this message]
2026-11-17 23:29       ` Bart Van Assche

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=cfbce18c-2a13-48ca-b2ac-938ad40ae892@kernel.dk \
    --to=axboe@kernel.dk \
    --cc=bvanassche@acm.org \
    --cc=chenaibing@cmss.chinamobile.com \
    --cc=edumazet@kernel.org \
    --cc=josef@toxicpanda.com \
    --cc=linux-block@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