View source for Mod:Hunt Research Group/ADF sricpt

Below is the script currently being employed for submitting the ADF jobs.

The jobs should be named NAME.com. The .com isn't needed for ADF, but it allows the submission script to work.

===Script===
<pre>
#!/bin/sh

# submit jobs to the que with this script using the following command:
# adf_sub is this script
# jobname is a name you will see in the qstat command
# name is the actual file minus .com etc it is passed into this script as ${in}
#
# qsub -N jobname -v in=name ~/bin/adf_sub

# batch processing commands
#PBS -l walltime=119:59:00
#PBS -l select=1:ncpus=4:mem=8GB
#PBS -j oe
#PBS -q pqph
#PBS -m ae

# load modules
#
module load adf/2016.106

#
# run adf
#
  adf {-n 4proc} < $PBS_O_WORKDIR/${in}.com> ${in}.log
  mkdir $PBS_O_WORKDIR/${in}
  cp $TMPDIR/$TAPE* /$PBS_O_WORKDIR/${in}/.
  cp $TMPDIR/$logfile /$PBS_O_WORKDIR/${in}/.
  cp $TMPDIR/${in} /$PBS_O_WORKDIR/${in}/.
  cp $TMPDIR/$t21* /$PBS_O_WORKDIR/${in}/.
  cp $TMPDIR/${in}.log /$PBS_O_WORKDIR/${in}/.
#  test -r $TMPDIR/fort.7
#  if [ $? -eq 0 ]
#  then
#    cp $TMPDIR/fort.7 /$PBS_O_WORKDIR/${in}.mos
#  fi
# exit
</pre>

This script is currently only running jobs on 1 processor. I am working on getting it running on more than 1 processor.

To run the script type "qsub -N jobname -v in=name ~/bin/adf_sub"

Jobname is the name that will be seen within the queue

name is the file name (no .com)