From mboxrd@z Thu Jan  1 00:00:00 1970
From: Hitoshi Mitake <mitake.hitoshi@gmail.com>
Subject: Re: [PATCH RFC v2] tgtd: send/recv iSCSI PDUs in worker threads
Date: Mon, 18 Nov 2013 00:46:42 +0900
Message-ID: <87vbzr81zx.wl%mitake.hitoshi@gmail.com>
References: <1384073347-3067-1-git-send-email-mitake.hitoshi@gmail.com>
	<20131117.225058.1568838937885772695.fujita.tomonori@lab.ntt.co.jp>
Mime-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka")
Return-path: <stgt-owner@vger.kernel.org>
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
        d=gmail.com; s=20120113;
        h=date:message-id:from:to:cc:subject:in-reply-to:references
         :user-agent:mime-version:content-type;
        bh=0iKi0icXwJN33dkvfdF4f9gQ6DSvL27Om5jxJvAxAfg=;
        b=pKGfWvtNEP6t69DNy2u8aYpJ80SDFagzpI4oZPnSucM52Wg1xgzqXjZOfmUUh0Q2Fu
         wltispktHgqT9nblJsJzkgkDd8PRf5xXg2BidZwCXWMzBa0qVGm8rRFHtDZkvAJx0G+/
         7zKm8E32/VZiGZzYK+7n9kOo+Jc+URuPs/5QNVj9XSq8otJKKIcTsSgxdZraVxkrtgBW
         TC1+RGZ+NAj4Ymvb4v9iVSGKmk4DQvCBaCqHqrl/uhZ7DjQ7OyKOA5R8Sm8+LilIPAYr
         vWhAnyQjiHq6oM/zrz0c51EjnFCWOMj5yVo5efFX4j2DeKLRxAT2PBTI70DNKa4aVZgS
         9xDA==
In-Reply-To: <20131117.225058.1568838937885772695.fujita.tomonori@lab.ntt.co.jp>
Sender: stgt-owner@vger.kernel.org
List-ID: <stgt.vger.kernel.org>
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: mitake.hitoshi@gmail.com, stgt@vger.kernel.org, mitake.hitoshi@lab.ntt.co.jp

At Sun, 17 Nov 2013 22:50:58 +0900 (JST),
FUJITA Tomonori wrote:
> 
> On Sun, 10 Nov 2013 17:49:07 +0900
> Hitoshi Mitake <mitake.hitoshi@gmail.com> wrote:
> 
> > From: Hitoshi Mitake <mitake.hitoshi@lab.ntt.co.jp>
> > 
> > Current tgtd sends and receives iSCSI PDUs in its main event
> > loop. This design can cause bottleneck when many iSCSI clients connect
> > to single tgtd process. For example, we need multiple tgtd processes
> > for utilizing fast network like 10 GbE because typical single
> > processor core isn't fast enough for processing bunch of requests.
> > 
> > This patch lets tgtd send/receive iSCSI PDUs and check digests in its
> > worker threads. After applying this patch, the bottleneck in the main
> > event loop is removed and the performance is improved.
> > 
> > The improvement can be seen even if tgtd and iSCSI initiator are
> > running on a single host. Below is a snippet of fio result on my
> > laptop. The workload is 128MB random RW. Backingstore is sheepdog.
> > 
> > Original tgtd:
> >   read : io=65392KB, bw=4445.2KB/s, iops=1111, runt= 14711msec
> >   write: io=65680KB, bw=4464.8KB/s, iops=1116, runt= 14711msec
> > 
> > tgtd with this patch:
> >   read : io=65392KB, bw=5098.9KB/s, iops=1274, runt= 12825msec
> >   write: io=65680KB, bw=5121.3KB/s, iops=1280, runt= 12825msec
> 
> Can you check how this scales with tcp connections on a
> reasonably-high-performance server?

Sure. I'd like to evaluate and post a result in the near future.

Thanks,
Hitoshi