All of lore.kernel.org
 help / color / mirror / Atom feed
From: Keir Fraser <keir.fraser@eu.citrix.com>
To: "Yang, Xiaowei" <xiaowei.yang@intel.com>, xen-devel@lists.xensource.com
Subject: Re: Memory allocation in NUMA system
Date: Fri, 25 Jul 2008 07:53:33 +0100	[thread overview]
Message-ID: <C4AF36FD.1B8CF%keir.fraser@eu.citrix.com> (raw)
In-Reply-To: <488949B3.7080205@intel.com>

On 25/7/08 04:34, "Yang, Xiaowei" <xiaowei.yang@intel.com> wrote:

> Let's say we have a 2-node system, with node0 and node1's memory range
> being 0-0xc0000000 (<4G) and 0x100000000-0x1c0000000 (>4G) respectively.
> In that case, node1's memory is always preferred for domain memory
> allocation, no matter which node the created domain is pinned to. It
> results in performance penalty.
> 
> One possible fix is to specify all range for the domain memory
> allocation, which means local memory is preferred. This change may be
> restricted only to the domain pinned to one node for less impact.
> 
> One side effect is that the DMA memory size may be smaller, which makes
> device domain unhappy. This can be addressed by reserving node0 to be
> used lastly.

Doesn't your solution amount to what we already do, for the 2-node example?
i.e., node0 would not be chosen until node1 is exhausted?

 -- Keir

  reply	other threads:[~2008-07-25  6:53 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-07-25  3:34 Memory allocation in NUMA system Yang, Xiaowei
2008-07-25  6:53 ` Keir Fraser [this message]
2008-07-25  7:22   ` Yang, Xiaowei
2008-07-25  7:27     ` Keir Fraser
2008-07-25  7:51       ` Yang, Xiaowei
2008-07-25  7:55         ` Keir Fraser
2008-07-25 10:26           ` Yang, Xiaowei
2008-07-25 12:56             ` Keir Fraser
2008-07-28 12:21               ` Andre Przywara
2008-07-28 12:38                 ` Keir Fraser
2008-07-28 14:26                   ` Andre Przywara
2008-07-28 14:53                     ` Keir Fraser

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=C4AF36FD.1B8CF%keir.fraser@eu.citrix.com \
    --to=keir.fraser@eu.citrix.com \
    --cc=xen-devel@lists.xensource.com \
    --cc=xiaowei.yang@intel.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.