Using Grid Engine

By specifying -batchsystem ge on the command line, Simcenter STAR-CCM+ automatically distributes processes to all CPUs/cores allocated to the job.

General information about job submission with batch systems can be found in Working in Unix-Based Batch Systems.

Please also refer to the specific cluster documentation and the batch system documentation for additional and required parameters or settings.

Documentation Reference

Useful Commands for Job Submission and Monitoring

Display SGE Release Version

$ qstat -v
OGS/GE 2011.11p1

Submit a Job

$ qsub <submission options> <script>

Show Job Status

$ qstat

Terminate Job

$ qdel <jobid>

Job Submission

A minimal, example jobscript to launch Simcenter STAR-CCM+ and automatically extract the information about allocated resources:

#!/bin/bash
 
#$ -N starsim
#$ -h_rt 1:00:00
#$ -pe mpi 32
#$ -o $JOB_NAME_$TASK_ID.out
#$ -e $JOB_NAME_$TASK_ID.err
 
STAR_PATH=/path/to/star/bin
$STAR_PATH/starccm+ -bs slurm -batch step my.sim

This allocates a job named starsim for one hour on 32 cores. The standard and error output is redirected to separate files. The allocated resources have to be specified on the command line, as denoted above. This example works only when there is a parallel environment defined named mpi.

Job File Submission Prefix

The directive prefix consists of two ASCII characters which, when appearing in the first two bytes of a script line, indicate that what follows is a Sun Grid Engine command. The default is "#$".

Submission Options


Parameter	Description
`-N <job_name>`	job name
`-l h_rt <hh:mm:ss>`	walltime
`-cwd`	run the job in the current working directory
`-pe <pename> <numcores>`	name of the physical environment to use for the job
`-q <queuename>`	assign the job to queue `<queuename>`
`-o <output_filename>`	output file
`-e <error_output_filename>`	error file

Submission Environment Variables


Environment Variable	Description
`JOB_ID`	job ID
`JOB_NAME`	name of the job
`NHOSTS`	number of hosts assigned to the job
`NSLOTS`	number of cores assigned to the job
`PE`	name of the Parallel environment
`PE_HOSTFILE`	job hostfile name
`QUEUE`	queue assigned to the job

Parallel Environment

A number of problems can occur through the use of a parallel environment that is not suited to the job. Avoid parallel environments using rsh unless your are specifically using rsh for internode connections.

$ qconf -spl
$ qconf -sp <PE>