[GE users] np_load_avg too low

Roch Belisle roch at island.ca
Tue Dec 20 19:02:47 GMT 2005



It's running on linux...


On Tue, Dec 20, 2005 at 02:00:07PM -0500, Rayson Ho wrote:
> Linux or HP-UX??
> 
> IIRC, the way that HP-UX calculates the load average is different than
> other OSes...
> 
> Rayson
> 
> 
> 
> On 12/20/05, Roch Belisle <roch at island.ca> wrote:
> > Hello,
> >
> > I'm currently having problems with the load average on 2-cpu Itanium
> > nodes.  Whether I submit jobs for 1 or 2 CPU's the load average never
> > goes above 1.  This is understandable for a single cpu machine but when
> > I have jobs running on 2 cpu's the load average should be much
> > higher (~2).  This also prevents me from queueing other jobs if the load average is
> > > 1.  How can I change the "max" load average so that my nodes can get
> > utilized to the fullest when submitting non-parallel jobs.
> >
> > Thanks
> >
> > Roch
> >
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
> > For additional commands, e-mail: users-help at gridengine.sunsource.net
> >
> >
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
> For additional commands, e-mail: users-help at gridengine.sunsource.net
> 

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net




More information about the gridengine-users mailing list