On Feb 27, 2012 11:39 PM, "Jeff Sturm" <jeff.sturm@xxxxxxxxxx> wrote:
>
> > -----Original Message-----
> > From: xen-users-bounces@xxxxxxxxxxxxx [mailto:xen-users-bounces@xxxxxxxxxxxxx] On
> > Behalf Of mywildimagination@xxxxxxxxx
> > Sent: Monday, February 27, 2012 9:53 AM
> >
> > Name                   ÂID Mem(MiB) VCPUs State  Time(s)
> > Domain-0 Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â 0 Â Â Â405 Â Â 2 r----- 778902.7
> > saturn                   1   7599   2 r-----
>
> Looks like the hardware has 2 CPU's available and 8GB RAM? ÂIs that right?
>
> If there are more CPUs available you can increase VCPUs for "saturn", likewise for memory. ÂOtherwise, you've given your domU all the resources available. ÂThe dom0 is lightly loaded, but that's normal. ÂYou can see from xentop that the system is busy.
>
> If the hardware isn't enough for the workload, you could build a new dom0 on a bigger host, and migrate the domU over.
>
> -Jeff
>
Well, yes your guest VM a.k.a domU (please don't use the term VPS) is overloaded... but that still depends on whether it's I/O-bound or CPU-bound...
The loads of domU's have no relation to the load of dom0; they are separate VMs. dom0 is not running the VMs. It's the hypervisor that's running the VMs (dom0 + all domU's) and the load if each VM is local to that particular VM.
In your case, dom0 is lightly-loaded (as it should be), while the single domU is heavily loaded. Check first what's causing such heavy load: not enough RAM making the system swap-thrashing? Overzealous polling by spinning? Buggy kernel?
Rgds,
