You will be requesting your code to run on a compute node through a Queue system, in this case the PBS (portable batch system). Our clusters have many queues available that differ based on the maximum time limit available to your code. The queue name has to be specified in the pbs script file, as shown on the Useful Scripts page.
To find the list of queues available (and their details) on the clusters, use the following command.
qstat -q
This command shows the list of all queues available.
Memory: maximum memory available on the queue
Walltime: maximum time available to your code on the given queue; the code will be killed after that time. Make sure that your code is genrating checkpoints at regular intervals to avoid data and time loss.
Run, Que and Lm denote the number of jobs running currently on the queue, number of queued jobs, and the maximum number of jobs allowed to run in the given queue.
Please note tha the private queues starting with the priv keyword may not allow you access to them, unless you have been given specific access. There are private queues associated with the schools, for instance, privse is a private queue for the school of engineering. You may request access to these to teh administrator.
Figure: A snapshot of qstat -q on the CPUHPC cluster
For more
detailed queue configuration specifications, use the qmgr -c “p s”
Please note that a low
priority queue lowpr is installed to utilize idle state of private
nodes on experimental basis , however
if owners of the private nodes are using their private queues it may a very long time for jobs to complete in this
queue as your job will be suspended once they demand access. While jobs are submitted in the private
queues, the job running in the private
nodes in lowpr, will go
to suspend state and the job will resume after the high priority jobs are completed. If due to any reason working of
private queues is affected by the jobs in lowpr
then it will be deleted with out
prior notice. Therefore, submission of jobs to this queue is at your own risk.
See the qstat manual page (man qstat) and Torque
official documentation for further details
Figure: qstat -q run on the GPUHPC cluster