All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andi Kleen <ak@suse.de>
To: "Martin J. Bligh" <mbligh@aracnet.com>
Cc: Brent Casavant <bcasavan@sgi.com>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org,
	hugh@veritas.com, ak@suse.de
Subject: Re: [PATCH] Use MPOL_INTERLEAVE for tmpfs files
Date: Tue, 2 Nov 2004 16:55:07 +0100	[thread overview]
Message-ID: <20041102155507.GA323@wotan.suse.de> (raw)
In-Reply-To: <14340000.1099410418@[10.10.2.4]>

On Tue, Nov 02, 2004 at 07:46:59AM -0800, Martin J. Bligh wrote:
> > This patch causes memory allocation for tmpfs files to be distributed
> > evenly across NUMA machines.  In most circumstances today, tmpfs files
> > will be allocated on the same node as the task writing to the file.
> > In many cases, particularly when large files are created, or a large
> > number of files are created by a single task, this leads to a severe
> > imbalance in free memory amongst nodes.  This patch corrects that
> > situation.
> 
> Yeah, but it also ruins your locality of reference (in a NUMA sense). 
> Not convinced that's a good idea. You're guaranteeing universally consistent
> worse-case performance for everyone. And you're only looking at a situation
> where there's one allocator on the system, and that's imbalanced.
> 
> You WANT your data to be local. That's the whole idea.

I think it depends on how you use tmpfs. When you use it for read/write
it's a good idea because you likely don't care about a bit of additional
latency and it's better to not fill up your local nodes with temporary
files.

If you use it with mmap then you likely want local policy.

But that's a big ugly to distingush, that is why I suggested the sysctl.

-Andi

WARNING: multiple messages have this Message-ID (diff)
From: Andi Kleen <ak@suse.de>
To: "Martin J. Bligh" <mbligh@aracnet.com>
Cc: Brent Casavant <bcasavan@sgi.com>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org,
	hugh@veritas.com, ak@suse.de
Subject: Re: [PATCH] Use MPOL_INTERLEAVE for tmpfs files
Date: Tue, 2 Nov 2004 16:55:07 +0100	[thread overview]
Message-ID: <20041102155507.GA323@wotan.suse.de> (raw)
In-Reply-To: <14340000.1099410418@[10.10.2.4]>

On Tue, Nov 02, 2004 at 07:46:59AM -0800, Martin J. Bligh wrote:
> > This patch causes memory allocation for tmpfs files to be distributed
> > evenly across NUMA machines.  In most circumstances today, tmpfs files
> > will be allocated on the same node as the task writing to the file.
> > In many cases, particularly when large files are created, or a large
> > number of files are created by a single task, this leads to a severe
> > imbalance in free memory amongst nodes.  This patch corrects that
> > situation.
> 
> Yeah, but it also ruins your locality of reference (in a NUMA sense). 
> Not convinced that's a good idea. You're guaranteeing universally consistent
> worse-case performance for everyone. And you're only looking at a situation
> where there's one allocator on the system, and that's imbalanced.
> 
> You WANT your data to be local. That's the whole idea.

I think it depends on how you use tmpfs. When you use it for read/write
it's a good idea because you likely don't care about a bit of additional
latency and it's better to not fill up your local nodes with temporary
files.

If you use it with mmap then you likely want local policy.

But that's a big ugly to distingush, that is why I suggested the sysctl.

-Andi
--
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:"aart@kvack.org"> aart@kvack.org </a>

  reply	other threads:[~2004-11-02 16:08 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-11-02  1:07 [PATCH] Use MPOL_INTERLEAVE for tmpfs files Brent Casavant
2004-11-02  1:07 ` Brent Casavant
2004-11-02  1:43 ` Dave Hansen
2004-11-02  1:43   ` Dave Hansen
2004-11-02  9:13 ` Andi Kleen
2004-11-02  9:13   ` Andi Kleen
2004-11-02 15:46 ` Martin J. Bligh
2004-11-02 15:46   ` Martin J. Bligh
2004-11-02 15:55   ` Andi Kleen [this message]
2004-11-02 15:55     ` Andi Kleen
2004-11-02 16:55     ` Martin J. Bligh
2004-11-02 16:55       ` Martin J. Bligh
2004-11-02 22:17       ` Brent Casavant
2004-11-02 22:17         ` Brent Casavant
2004-11-02 22:51         ` Martin J. Bligh
2004-11-02 22:51           ` Martin J. Bligh
2004-11-03  1:12           ` Brent Casavant
2004-11-03  1:12             ` Brent Casavant
2004-11-03  1:30             ` Martin J. Bligh
2004-11-03  8:44           ` Hugh Dickins
2004-11-03  8:44             ` Hugh Dickins
2004-11-03  9:01             ` Andi Kleen
2004-11-03  9:01               ` Andi Kleen
2004-11-03 16:32               ` Brent Casavant
2004-11-03 16:32                 ` Brent Casavant
2004-11-03 21:00                 ` Martin J. Bligh
2004-11-03 21:00                   ` Martin J. Bligh
2004-11-08 19:58                   ` Brent Casavant
2004-11-08 19:58                     ` Brent Casavant
2004-11-08 20:57                     ` Martin J. Bligh
2004-11-08 20:57                       ` Martin J. Bligh
2004-11-09 19:04                     ` Hugh Dickins
2004-11-09 19:04                       ` Hugh Dickins
2004-11-09 20:09                       ` Martin J. Bligh
2004-11-09 20:09                         ` Martin J. Bligh
2004-11-09 21:08                         ` Hugh Dickins
2004-11-09 21:08                           ` Hugh Dickins
2004-11-09 22:07                           ` Martin J. Bligh
2004-11-09 22:07                             ` Martin J. Bligh
2004-11-10  2:41                       ` Brent Casavant
2004-11-10  2:41                         ` Brent Casavant
2004-11-10 14:20                         ` Hugh Dickins
2004-11-10 14:20                           ` Hugh Dickins
2004-11-11 19:48                       ` Hugh Dickins
2004-11-11 19:48                         ` Hugh Dickins
2004-11-11 23:10                         ` Brent Casavant
2004-11-11 23:10                           ` Brent Casavant
2004-11-15 22:07                         ` Brent Casavant
2004-11-15 22:07                           ` Brent Casavant

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=20041102155507.GA323@wotan.suse.de \
    --to=ak@suse.de \
    --cc=bcasavan@sgi.com \
    --cc=hugh@veritas.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mbligh@aracnet.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.