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] Slow disk IO

To: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-users] Slow disk IO
From: Mike McGrath <mmcgrath@xxxxxxxxxx>
Date: Sun, 28 Dec 2008 13:39:10 -0600 (CST)
Delivery-date: Sun, 28 Dec 2008 11:39:53 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
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>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Alpine 2.00 (LFD 1167 2008-08-23)
I'm seeing slower then expected disk IO.  Also, using sar, it seems the
disk is measuring different utilization on the domU then the dom0.  I've
attached to the disk with:

"phy:/dev/PowerVault00/KojiShare,xvdb1,w"

That logical volume has a physical volume of an attached 15 disk raid6
array (an MD1000 from dell).

Here is the sar data I'm referencing:

http://mmcgrath.fedorapeople.org/dom0

and

http://mmcgrath.fedorapeople.org/domU

The disk does have one other guest attached to it but that guest has very
little disk IO.

Other items of note: the data periods above happened during a bacula
backup.  I would have expected much higher data throughput.  We're talking
about an average read speed of 3.58 MBytes/s and it says its 95% utilized.
Meanwhile the dom0 says its only 51% utilized.

Additionally on both the domU and dom0 we're not seeing much iowait.  On
the domU we're only talking about an average of 14%.  If the disks truly
were fully utilized, I'd expect that to be much higher. These are both
RHEL5 boxes.  dom0=2.6.18-92.1.18.el5xen.  domU=2.6.18-105.el5xen.  Also
both are x86_64.

Thoughts?

        -Mike

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

<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-users] Slow disk IO, Mike McGrath <=