Nodes belonging to the birthright, chem, cnms, ccsd, bsd, mstd, ntrc, nucphys,
virtues, theory, qmc, and ccsi groups have now moved to the Slurm scheduler
and can be accessed from
See the Running Jobs documentation for more details.
SHPC utilizes Torque/Moab to manage jobs that users submit to various queues on a computer system. Each queue represents a group of resources with attributes necessary for the queue's jobs. You can see the list of queues that SHPC has by typing
qstat -q. batch is the default queue.
📝 Note: Do not run jobs on the login nodes. All jobs launched from those nodes will be terminated without notice.
To list all jobs:
To refine the list of jobs to only those submitted by a user:
qstat -u username
To further refine the list of jobs, the following command will list jobs submitted by a user and which are running.
qstat -u username -s -r
To obtain the status of a job, run the following command using the job's ID number (this is provided at time of job submission).
qstat -f job_ID
You can also use
checkjob job_ID to show the current status of the job.
Submitting a job
To submit a job, use the
qsub command, followed by the name of your submission file. A Job ID will be provided. You may want to make note of the ID for later use.
Deleting a job
📝 Note: Be aware that deleting a job cannot be undone. Double check the job ID before deleting a job.
Users can delete their jobs by typing the following command.
To delete all the jobs of a user:
qdel $(qselect -u username)
By default, the Moab scheduler uses as few nodes as possible to schedule a job. If you have a job that requests two nodes with sixteen cores each, it will condense that request into one node with 32 cores. To override this default behaviour, you can use the following option with qsub:
qsub -W x=nmatchpolicy:exactnode your_script