[torqueusers] Need Suggestions for a qmgr script for multicore
cluster
Joshua Bernstein
jbernstein at penguincomputing.com
Wed Mar 18 12:27:25 MDT 2009
Hello Samir,
You'll want to make sure you that your /var/spool/torque/server_priv/nodes file
lists each node and includes the number of cores on that nodes, so for a dual
socket, quad core system the file should include the nodes hostname, followed by
np=8. For example:
node0 np=8
node1 np=8
...
node5 np=8
You will need to restart pbs_server to make any changes take effect. Also, you
can confirm the proper setting by using pbsnodes -a, and check to make sure
pbs_server has a np=8 line for each node.
-Joshua Bernstein
Software Engineer
Penguin Computing
Samir Khanal wrote:
> Hi All
>
> I have a compute cluster with 6 compute nodes with a quad core Intel in each.
> (total 24 cores and 6 nodes)
> I have configured Torque for single core computers previously.
>
> How is it different to configure a multicore computers with torque?
> I am using torque pbs_server version 2.3.6 that came with Rocks clusters.
>
> Basically i want to run MPI applications.
>
> Currently i have following q properties
>
> create queue batch
> set queue batch queue_type = Route
> set queue batch max_running = 24
> set queue batch route_destinations = serial
> set queue batch route_destinations += parallel
> set queue batch enabled = True
> set queue batch started = True
> #
> # Create and define queue default
> #
> create queue default
> set queue default queue_type = Execution
> set queue default enabled = True
> set queue default started = True
> #
> # Create and define queue parallel
> #
> create queue parallel
> set queue parallel queue_type = Execution
> set queue parallel Priority = 50
> set queue parallel resources_max.nodect =6
> set queue parallel resources_max.nodes = 6
> set queue parallel resources_max.walltime = 48:00:00
> set queue parallel resources_min.nodect = 1
> set queue parallel resources_min.nodes = 1
> set queue parallel resources_default.nodect = 2
> set queue parallel resources_default.nodes = 2
> set queue parallel resources_default.walltime = 01:00:00
> set queue parallel enabled = True
> set queue parallel started = True
>
> Could you please give me some suggestions so that i can utilize the cores and nodes properly.
>
> Thanks
> Samir
> _______________________________________________
> torqueusers mailing list
> torqueusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torqueusers
More information about the torqueusers
mailing list