[torqueusers] simulating pbs/torque

Ken Nielson knielson at clusterresources.com
Tue Jun 23 09:47:01 MDT 2009


Paul,

We have developed a torque which will allow multiple moms to run on the 
same node similar to what you have described. It is built on the 2.3.7 
code base for testing purposes. Would you be interested in trying out 
what we have?

Ken Nielson
Cluster Resources, Inc.

Paul D Marshall wrote:
> Hello,
>
> I was wondering if someone would be able to point me in the right direction.
>
> I'm working with Torque 2.3.6 and I've been adding and removing machines 
> from my cluster dynamically (just via the qmgr command and starting or 
> killing pbs_mom processes on the workers).
>
> I only have so many machines I can actually work with and so I've begun 
> to look into simulating the pbs_mom process. The problem that I 
> encountered with this is that there seem to be a variety of reasonably 
> complicated messages exchanged between the pbs_mom and server. So before 
> I dug myself too deep, I thought I'd check with others.
>
> Does anyone know of another project or person who has simulated torque 
> on *some level*?
>
> If not, can anyone provide suggestions as to where might be an 
> appropriate point to simulate the dispatching, running, and reporting of 
> jobs on pbs workers?
>
> Thanks,
> Paul
> _______________________________________________
> torqueusers mailing list
> torqueusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torqueusers
>   



More information about the torqueusers mailing list