WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-changelog

[Xen-changelog] [xen-unstable] xen: turn numa=on by default

To: xen-changelog@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-changelog] [xen-unstable] xen: turn numa=on by default
From: Xen patchbot-unstable <patchbot-unstable@xxxxxxxxxxxxxxxxxxx>
Date: Tue, 01 Dec 2009 06:30:51 -0800
Delivery-date: Tue, 01 Dec 2009 06:33:08 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-changelog-request@lists.xensource.com?subject=help>
List-id: BK change log <xen-changelog.lists.xensource.com>
List-post: <mailto:xen-changelog@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-changelog>, <mailto:xen-changelog-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-changelog>, <mailto:xen-changelog-request@lists.xensource.com?subject=unsubscribe>
Reply-to: xen-devel@xxxxxxxxxxxxxxxxxxx
Sender: xen-changelog-bounces@xxxxxxxxxxxxxxxxxxx
# HG changeset patch
# User Keir Fraser <keir.fraser@xxxxxxxxxx>
# Date 1259675987 0
# Node ID 939933401650d5f5e221cfa8005d3872b16ef8f0
# Parent  cc7d66ba0dad4147a30457c5d5408ec88c60daa4
xen: turn numa=on by default

I did some benchmark runs (lmbench & kernel compile) with a number of
guests running in parallel to compare the performance of numa=on vs.
numa=off.  As soon as one starts to load the machine, the performance
goes down in the numa=off case.  The tests were done on an 8-node
machine (4 cores each).  lmbench (actually copying large amounts of
memory) shows a dramatic dropdown, but I even noticed significant
performance decrease for a tmpfs based Linux kernel compile. Here a
summary of the data:

lmbench's rd benchmark (normalized to native Linux (=100)):
guests    numa=off       numa=on       avg increase
        min  avg  max  min  avg  max
     1       78.0           102.3
     7  37.4 45.6 62.0 90.6 102.3 110.9 124.4%
    15  21.0 25.8 31.7 41.7 48.7 54.1    88.2%
    23  13.4 17.5 23.2 25.0 28.0 30.1    60.2%

kernel compile in tmpfs, 1 VCPU, 2GB RAM, average of elapsed time:
guests    numa=off   numa=on   increase
      1    480.610    464.320    3.4%
      7    482.109    461.721    4.2%
     15    515.297    477.669    7.3%
     23    548.427    495.180    9.7%
again with 2 VCPUs and make -j2:
      1    264.580    261.690    1.1%
      7    279.763    258.907    7.7%
     15    330.385    272.762   17.4%
     23    463.510    390.547   15.7% (46 VCPUs on 32pCPUs)

Selected tests on a 4-node machine showed similar behavior (7.9 %
increase with 6 parallel guests on the 2 VCPU kernel compile
benchmark).

Note that this does not affect non-NUMA machines at all, since NUMA
will be turned off again by the code if no NUMA topology is detected.

Signed-off-by: Andre Przywara <andre.przywara@xxxxxxx>
---
 xen/arch/x86/numa.c |    3 +--
 1 files changed, 1 insertion(+), 2 deletions(-)

diff -r cc7d66ba0dad -r 939933401650 xen/arch/x86/numa.c
--- a/xen/arch/x86/numa.c       Tue Dec 01 13:57:02 2009 +0000
+++ b/xen/arch/x86/numa.c       Tue Dec 01 13:59:47 2009 +0000
@@ -41,8 +41,7 @@ cpumask_t node_to_cpumask[MAX_NUMNODES] 
 
 nodemask_t __read_mostly node_online_map = { { [0] = 1UL } };
 
-/* Default NUMA to off for now. acpi=on required to enable it. */
-int numa_off __devinitdata = 1;
+int numa_off __devinitdata = 0;
 
 int acpi_numa __devinitdata;
 

_______________________________________________
Xen-changelog mailing list
Xen-changelog@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-changelog

<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-changelog] [xen-unstable] xen: turn numa=on by default, Xen patchbot-unstable <=