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

Re: [Xen-users] Managing Dom0 load?

To: Xen User-List <xen-users@xxxxxxxxxxxxxxxxxxx>
Subject: Re: [Xen-users] Managing Dom0 load?
From: Berni Elbourn <berni@xxxxxxxxxxxxxxxxxxxx>
Date: Wed, 03 Jun 2009 11:32:42 +0100
Delivery-date: Wed, 03 Jun 2009 03:33:23 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <7207d96f0906030138q72b00d3dkc0f9dedc2d0ee7f4@xxxxxxxxxxxxxx>
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/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=unsubscribe>
References: <4A2633F2.9010607@xxxxxxxxxxxxxxxxxxxx> <7207d96f0906030138q72b00d3dkc0f9dedc2d0ee7f4@xxxxxxxxxxxxxx>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mozilla-Thunderbird 2.0.0.19 (X11/20090103)
Fajar A. Nugraha wrote:
On Wed, Jun 3, 2009 at 3:27 PM, Berni Elbourn
<berni@xxxxxxxxxxxxxxxxxxxx> wrote:
I am using Debian lenny in Dom0. Disk is a two disk md raid1. DomUs have
physical lvm.

It seems no matter how I assign the affinity in the DomUs:- top on Dom0
shows cpu0 is "clogged up" with typically 0%id 80%wa yet cores 1-3 are
99.7%id. eg:

Most likely you're I/O bound. Try:
- iostat -x 3, see which disk/block device are busy
- xm top. Should show you which dom uses most I/O (works for PV guest
or HVM with PV drivers)


Hi Fajar, Yes yes the load is caused by using stress in one of the domu using -hdd parameter but it also has -cpu.

No matter the cpu configuration it seems most of the load goes on the first core.

I am somewhat curious if and how the workload can be spread about more evenly on the cores?

Berni

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

<Prev in Thread] Current Thread [Next in Thread>