Message-ID: <1431090801.1418.87.camel@sauron.fi.intel.com>
Subject: Re: [PATCH] numa,sched: only consider less busy nodes as numa
 balancing destination
From: Artem Bityutskiy <dedekind1@gmail.com>
Reply-To: dedekind1@gmail.com
To: Rik van Riel <riel@redhat.com>
Cc: linux-kernel@vger.kernel.org, mgorman@suse.de, peterz@infradead.org,
        jhladky@redhat.com
Date: Fri, 08 May 2015 16:13:21 +0300
In-Reply-To: <20150506114128.0c846a37@cuia.bos.redhat.com>
References: <1430908530.7444.145.camel@sauron.fi.intel.com>
	 <20150506114128.0c846a37@cuia.bos.redhat.com>
Content-Type: text/plain; charset="UTF-8"
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1777
Lines: 50

On Wed, 2015-05-06 at 11:41 -0400, Rik van Riel wrote:
> On Wed, 06 May 2015 13:35:30 +0300
> Artem Bityutskiy <dedekind1@gmail.com> wrote:
> 
> > we observe a tremendous regression between kernel version 3.16 and 3.17
> > (and up), and I've bisected it to this commit:
> > 
> > a43455a sched/numa: Ensure task_numa_migrate() checks the preferred node
> 
> Artem, Jirka, does this patch fix (or at least improve) the issues you
> have been seeing?  Does it introduce any new regressions?

Hi Rik,

first of all thanks for your help!

I've tried this patch and it has very small effect. I've also ran the
benchmark with auto-NUMA disabled too, which is useful, I think. I used
the tip of Linuses tree (v4.1-rc2+).


 Kernel         Avg response time, ms
------------------------------------------------------
Vanilla                1481
Patched                1240
Reverted               256
Disabled               309


Vanilla: pristine v4.1-rc2+
Patched: Vanilla + this patch
Reverted: Vanilla + a revert of a43455a
Disabled: Vanilla and auto-NUMA disabled via procfs

I ran the benchmark for 1 hour for every configuration this time. I
cannot say for sure the deviation right now, but I think it is tens of
milliseconds, so disabled vs reverted _may_ be within the error range,
but I need to do more experiments.

So this patch dropped the average Web server response time dropped from
about 1.4 seconds to about 1.2 seconds, which isn't a bad improvement,
but it is far less than what we get when reverting that patch.

Artem.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/