linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Wu Fengguang <fengguang.wu@intel.com>
To: Chris Webb <chris@arachsys.com>
Cc: Minchan Kim <minchan.kim@gmail.com>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
	Pekka Enberg <penberg@cs.helsinki.fi>,
	Andi Kleen <andi@firstfloor.org>,
	Lee Schermerhorn <lee.schermerhorn@hp.com>,
	Christoph Lameter <cl@linux-foundation.org>
Subject: Re: Over-eager swapping
Date: Wed, 18 Aug 2010 23:21:03 +0800	[thread overview]
Message-ID: <20100818152103.GA11268@localhost> (raw)
In-Reply-To: <20100818144655.GX2370@arachsys.com>

Andi, Christoph and Lee:

This looks like an "unbalanced NUMA memory usage leading to premature
swapping" problem.

Thanks,
Fengguang

On Wed, Aug 18, 2010 at 10:46:59PM +0800, Chris Webb wrote:
> Wu Fengguang <fengguang.wu@intel.com> writes:
> 
> > Did you enable any NUMA policy? That could start swapping even if
> > there are lots of free pages in some nodes.
> 
> Hi. Thanks for the follow-up. We haven't done any configuration or tuning of
> NUMA behaviour, but NUMA support is definitely compiled into the kernel:
> 
>   # zgrep NUMA /proc/config.gz 
>   CONFIG_NUMA_IRQ_DESC=y
>   CONFIG_NUMA=y
>   CONFIG_K8_NUMA=y
>   CONFIG_X86_64_ACPI_NUMA=y
>   # CONFIG_NUMA_EMU is not set
>   CONFIG_ACPI_NUMA=y
>   # grep -i numa /var/log/dmesg.boot 
>   NUMe: Allocated memnodemap from b000 - 1b540
>   NUMA: Using 20 for the hash shift.
> 
> > Are your free pages equally distributed over the nodes? Or limited to
> > some of the nodes? Try this command:
> > 
> >         grep MemFree /sys/devices/system/node/node*/meminfo
> 
> My worst-case machines current have swap completely turned off to make them
> usable for clients, but I have one machine which is about 3GB into swap with
> 8GB of buffers and 3GB free. This shows
> 
>   # grep MemFree /sys/devices/system/node/node*/meminfo
>   /sys/devices/system/node/node0/meminfo:Node 0 MemFree:          954500 kB
>   /sys/devices/system/node/node1/meminfo:Node 1 MemFree:         2374528 kB
> 
> I could definitely imagine that one of the nodes could have dipped down to
> zero in the past. I'll try enabling swap on one of our machines with the bad
> problem late tonight and repeat the experiment. The node meminfo on this box
> currently looks like
> 
>   # grep MemFree /sys/devices/system/node/node*/meminfo
>   /sys/devices/system/node/node0/meminfo:Node 0 MemFree:           82732 kB
>   /sys/devices/system/node/node1/meminfo:Node 1 MemFree:         1723896 kB
> 
> Best wishes,
> 
> Chris.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2010-08-18 15:21 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-02 12:47 Over-eager swapping Chris Webb
2010-08-02 23:55 ` Minchan Kim
2010-08-03  3:31   ` Chris Webb
2010-08-03  4:09     ` Minchan Kim
2010-08-03  4:28       ` Wu Fengguang
2010-08-03  4:47         ` Minchan Kim
2010-08-03  6:39           ` Wu Fengguang
2010-08-03 21:49         ` Chris Webb
2010-08-04  2:21           ` Wu Fengguang
2010-08-04  3:10             ` Minchan Kim
2010-08-04  3:24               ` Wu Fengguang
2010-08-04  9:58                 ` Chris Webb
2010-08-04 11:49                   ` Wu Fengguang
2010-08-04 12:04                     ` Chris Webb
2010-08-18 14:38                       ` Wu Fengguang
2010-08-18 14:46                         ` Chris Webb
2010-08-18 15:21                           ` Wu Fengguang [this message]
2010-08-18 15:57                             ` Christoph Lameter
2010-08-18 16:20                               ` Wu Fengguang
2010-08-18 15:57                             ` Lee Schermerhorn
2010-08-18 15:58                               ` Chris Webb
2010-08-18 16:13                                 ` Christoph Lameter
2010-08-18 16:32                                   ` Chris Webb
2010-08-19  5:16                                   ` Balbir Singh
2010-08-19 10:20                                   ` Chris Webb
2010-08-19 19:03                                     ` Christoph Lameter
2010-08-18 16:13                                 ` Wu Fengguang
2010-08-18 16:31                                   ` Chris Webb
2010-08-19  5:13         ` Balbir Singh
2010-08-18 16:45 ` Balbir Singh
2010-08-19  9:25   ` Chris Webb
2010-08-19 15:13     ` Balbir Singh
  -- strict thread matches above, loose matches on Subject: below --
2012-04-23  9:27 Richard Davies
2012-04-23 12:07 ` Zdenek Kaspar
2012-04-23 17:19 ` Dave Hansen
2012-04-24  0:35 ` Minchan Kim
2012-04-24 11:16 ` Peter Lieven
2012-04-25 14:41 ` Rik van Riel

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=20100818152103.GA11268@localhost \
    --to=fengguang.wu@intel.com \
    --cc=andi@firstfloor.org \
    --cc=chris@arachsys.com \
    --cc=cl@linux-foundation.org \
    --cc=kosaki.motohiro@jp.fujitsu.com \
    --cc=lee.schermerhorn@hp.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=minchan.kim@gmail.com \
    --cc=penberg@cs.helsinki.fi \
    /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).