Torque

Useful commands for submitting a job

To submit a job to the PBS queue and request additional resources use the qsub command. Status of a job already in the PBS queue can be checked with the qstat command. PBS script which contains the number of nodes and processors desired for the job, how much wall-clock time is being requested, program to execute and more can be used to simplify the process. Example settings to be put in pbs script:
To specify job name:

#PBS -N job_name

standard output file:

 #PBS -o stdout_file

error file:

 #PBS -e stderr_file

maximum cpu time and wall time. If the job does not finish by the time reached, the job is terminated.:

 #PBS -l cput=hhhh:mm:ss

 #PBS -l walltime=hhhh:mm:ss

number of nodes and processors per node:

 #PBS -l nodes=2:ppn=16

the maximum amount of physical memory required. Setting it too large can result in your job waiting in the queue for sufficient resources to become available.:
#PBS -l mem=4096mb

destination of the job queue and/or server. :

 #PBS -q queue@server

priority for the job from(integer from -1024 to 1023). The default is 0.:

 #PBS -p 0

Submitting the job:

qsub myjob.pbs

qsub accepts additional parameters to alter settings. For example if additional cpus are needed:
qsub -l nodes=4:ppn=2 myjob.pbs

To check for information:
qstat without arguments displays status for all jobs in the system.
For a specific job:

qstat jobID

For all of your own jobs:

qstat -u username

For additional information:

qstat -l jobID

To determine which nodes have been allocated to a given job, the user can run:

qstat –na

To view all jobs in a particular queue, use:

qstat queue_name

To view a summary of all queues use:

qstat -q