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/
Home Products Support Community News


[Xen-users] Credit Scheduler

To: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-users] Credit Scheduler
From: Florian Schmaus <flo4lists@xxxxxxxxxx>
Date: Mon, 01 Dec 2008 16:23:42 +0100
Delivery-date: Mon, 01 Dec 2008 07:22:34 -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: Thunderbird (X11/20081110)
I have a CentOS 5.2 Dual-Core Host with two domU's. One domU is for apache and postfix, the other one is just for BOINC. I thought that I can configure the XEN credit scheduler in such a manner, that the BOINC domU only gets CPU resources if the other one, the "important" one, has nothing else to do. After a "xm sched-credit -d boinc -w 1" and "xm sched-credit -d web -w 200" I run a CPU benchmark on the "web" domU. I expected that the "web" domU would get 199% CPU share and the boinc domU just 1%, because the "web" domU hast 200 times more weight than the BOINC domU. But the credit scheduler gave both domU's about 100% CPU share.
What am I missing here?

xm dmesg shows "(XEN) Using scheduler: SMP Credit Scheduler (credit)"
and "xm sched_credit -d web" also confirms that the weight is set correctly.


Xen-users mailing list

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