Input and Output¶
- The Moab job scheduler will assist you with the collection of program output on the various nodes depending upon the options selected when the job is submitted.
#PBS -k oeTincluded in the batch submit file requests the scheduler collect all output and error information and place them in your home directory with a name based upon the job name, the job ID and if it's standard or error output.
#PBS -N hello #PBS -l nodes=7:ppn=4 #PBS -l walltime=15:00:00 #PBS -q paceib #PBS -k oe #PBS -m abe cd ~/mpi echo "Started on `/bin/hostname`" echo "Nodes chosen are:" cat $PBS_NODEFILE /usr/bin/mpirun -v -np 28 -machinefile $PBS_NODEFILE ~/mpi/hello
- The above job will create, assuming it was assigned job number 244 when submitted, the following two files in your home directory, using the name hello from the job file:
- hello.e244 (error output from job #244)
- hello.o244 (standard job output from job #244)
- The collected output from all CPUs (28, in this example) will be in each appropriate file.
- If the
#PBS -k oeline is omitted or commented out in the script, then the output files will be created in the directory where the job was launched from.