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-users

[Xen-users] Does Driver Domain affects the performance of DomU?

To: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-users] Does Driver Domain affects the performance of DomU?
From: Jinseok Kim <shiefra82@xxxxxxxxxxx>
Date: Tue, 18 Sep 2007 23:31:20 -0700 (PDT)
Delivery-date: Tue, 18 Sep 2007 23:32:20 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-users-request@lists.xensource.com?subject=help>
List-id: Xen user discussion <xen-users.lists.xensource.com>
List-post: <mailto:xen-users@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
Hello.

Because DomUs use Driver-Domain to perform their I/O operations,
I expected if Driver-Domain is busy performing other jobs (not I/O operation
on behalf of DomUs),
DomUs would experience performance degradation.
So I constructed an experiment.

I have a server with two quad-core processors and then installed Xen 3.0.
Because there are 8 physical cores, I created one Dom0 and 7 DomUs and then
allocated 8 cores
to each VM one by one using 'xm vcpu-p' command.
And edited '/etc/xen/xend-config.sxp' so that made Dom0 use just 1 cpu.
(dom0-cpus 1)
And wrote an infinite loop program in C to make Dom0 busy.
(If I run the loop program on Dom0, the CPU usage of Dom0 increases up to
100% but it can't exceeds
100% even if I run two loop programs simultaneously.)

With this settings, I performed DB benchmark, sql-bench, on the 7 DomUs
simultaneously.
At first I ran the loop program on Dom0 and performed benchmarks,
and next I performed benchmarks again without the loop program.

I found there is no significant difference between two results.
According to my expectation, the benchmark result of former should be lower
than the result of latter.
But it wasn't.

Is my understand incorrect?
Or were there any mistakes in experimental setup?

Thanks.
-- 
View this message in context: 
http://www.nabble.com/Does-Driver-Domain-affects-the-performance-of-DomU--tf4479047.html#a12771559
Sent from the Xen - User mailing list archive at Nabble.com.


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

<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-users] Does Driver Domain affects the performance of DomU?, Jinseok Kim <=