[torqueusers] Installing Torque on shared disk

Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu
Sat Aug 13 23:47:53 MDT 2011


On Fri, Aug 12, 2011 at 3:07 PM, William Nelson <bnelson at arabidopsis.org> wrote:
> I'm new to Torque and have just been given the responsibility to support an
> existing cluster that uses Torque.
> This cluster has Torque installed on a shared disk with links from
> /usr/local to the shared TORQUE_HOME on the master and the nodes.
> Torque isn't used very often here because the users say that it is flakey -
> quits during production runs.

that has most likely other reasons.

> My simple test runs have shown it to be working fine though so I want to
> confirm that the installation is okay before I go deeper.
> Since the installation procedure that I have says to install Torque on each
> individual node I'm concerned that there could be concurrency issues when
> Torque is accessing shared resource files. Should I be concerned?

no. many people run torque of shared nfs exports. i run a 72 node
cluster that has currently an average uptime of over 510 days and
that has been running torque 2.3.6 off a shared nfs export almost
two years and served over 50,000 jobs in that period. there are likely
systems out there with even more impressive numbers.

cheers,
     axel.

> Thanks,
> Bill
>
>
>
> _______________________________________________
> torqueusers mailing list
> torqueusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torqueusers
>
>



-- 
Dr. Axel Kohlmeyer    akohlmey at gmail.com
http://sites.google.com/site/akohlmey/

Institute for Computational Molecular Science
Temple University, Philadelphia PA, USA.


More information about the torqueusers mailing list