Using the HPSS Queue
The HPSS queue can be used to transfer files or directories to hpss using a batch file. Jobs running in this queue are not allocated compute nodes, so the aprun command will fail and should not be added to batch files submitted to this queue on Kraken. Nautilus batch scripts do not use aprun and examples can be seen on the Nautilus batch script examples page. You may only submit jobs to the HPSS queue if you logged into a node with your RSA SecurID OTP token. The wall clock limit for the hpss queue is 24 hours. To submit jobs to the hpss queue, use the qsub -q option (#PBS -q hpss). Using job dependencies, you can schedule an HPSS job to stage data before and/or after a normal production job.
Following is an example of a simple HPSS batch file with a dependency.
#!/bin/bash #PBS -A TG-EXAMPLE #PBS -l walltime=10:00:00 #PBS -q hpss #PBS -W depend=afterok:123456 cd $PBS_O_WORKDIR hsi put file htar cvf this_run.tar dir/
This will simply be submitted using
> qsub filename
- #PBS -q hpss - submit jobs to the hpss queue
- #PBS -W depend=afterok:123456- stage the transfer after the completion of job 123456
- #PBS -l walltime=10:00:00 - The wall clock limit for the hpss queue is 24 hours.