| Both sides previous revision Previous revision Next revision | Previous revision |
| doku:testquickstart [2014/06/17 12:46] – [The job script] ir | doku:testquickstart [Unknown date] (current) – external edit (Unknown date) 127.0.0.1 |
|---|
| |
| ==== The job script ==== | ==== The job script ==== |
| Define name and number of slots: \\ <code>#$ -N <job_name> | === header === |
| | <code>#$ -N <job_name> |
| #$ -pe mpich <slots> | #$ -pe mpich <slots> |
| #$ -V</code> | #$ -V |
| where "<job_name>" is a freely chosen descriptive name and "<slots>" is the number of processor cores that you want to use for the calculation. To ensure exclusive reservation of the compute nodes for your job, | #$ -l h_rt=hh:mm:ss |
| the value for "<slots>" has to be a multiple of 8. | #$ -M <email address to notify of job events> |
| "-V" declares that all environment variables in the qsub command's environment are to be exported to the batch job. | #$ -m beas # all job events sent via email </code> |
| | - "<job_name>" is a freely chosen descriptive name, |
| | - "<slots>" is the number of processor cores that you want to use for the calculation. To ensure exclusive reservation of the compute nodes for your job, the value for "<slots>" has to be a multiple of 8. |
| | - "-V" declares that all environment variables in the qsub command's environment are to be exported to the batch job. |
| | - "-l" specifies the job's runtime. This explicit specification is in particular advisable for jobs with short run times, i.e., several hours or even minutes. In order to **reduce the time in the queue**, see also the section on [[doku:runtime|maximum runtime specification.]] |
| | - "-M <email address>; -m beas" request E-Mail notifications concerning job events (b .. beginning, e .. end, a .. abort or reschedule, s .. suspend). |
| |
| | === run executable === |
| The job can be started in several ways, | The job can be started in several ways, |
| - as single core job on one core (no MPI) task <code>./<executable></code> | - as single core job on one core (no MPI) task <code>./<executable></code> |
| mpirun -m $TMPDIR/machines -np $NSLOTS <executable></code> | mpirun -m $TMPDIR/machines -np $NSLOTS <executable></code> |
| | |
| "<executable>" is substituted by the path of the MPI-enabled application.\\ To receive E-Mail notifications concerning job events (b .. beginning, e .. end, a .. abort or reschedule, s .. suspend), use these lines in your job script: \\ <code>#$ -M <email address to notify of job events> | "<executable>" is substituted by the path of the MPI-enabled application. |
| #$ -m beas # all job events sent via email</code> It is often advisable to also specify the job's runtime as <code>#$ -l h_rt=hh:mm:ss</code> in particular when you know that your job will run only for several hours or even minutes. That way one can "backfill" the queue, thus avoiding very long waiting times, which may be due to a highly parallel job waiting for free resources. \\ Here is an example job-script, requesting 32 processor cores, which will run for a maximum of 3 hours and sends emails at the beginning and at the end of the job: <code> | |
| | |
| | ==== example ==== |
| | Here is an example job-script, requesting 32 processor cores, which will run for a maximum of 3 hours and sends emails at the beginning and at the end of the job: <code> |
| #$ -N hitchhiker | #$ -N hitchhiker |
| #$ -pe mpich 32 | #$ -pe mpich 32 |
| |
| |
| [[doku:runtime|In order to reduce the time in the queue, see the section on maximum runtime specification.]] | |
| |
| ====Submit your job:==== | ====Submit your job:==== |