Torque Examples

You may use the following Torque batch script as a template. Lines that begin with #PBS are parsed to define Torque options, they're not just comments.  More information about job submission can be found on the Torque Wiki.

#!/bin/sh
### Set the job name
#PBS -N myprogram

### Declare myprogram non-rerunable
#PBS -r n

### Uncomment to send email when the job is completed:
### #PBS -m ae
### #PBS -M your@email.address

### Optionally specifiy destinations for your myprogram's output
### Specify localhost and an NFS filesystem to prevent file copy errors.
### #PBS -e localhost:/work/username/test.err
### #PBS -o localhost:/work/username/test.log

### Optional similar syntax words might work for staging files.
### #PBS -W stagein=/scratch/file.in@localhost:/full/file.in
### #PBS -W stageout=/scratch/file.out@localhost:/full/file.out

### Set the queue to "batch", the only available queue. 
#PBS -q batch

### Specify the number of cpus for your job.  This example will run on 16 cpus 
### using 8 nodes with 2 processes per node.  
### You MUST specify some number of nodes or Torque will fail to load balance.
#PBS -l nodes=8:ppn=2

### You should tell PBS how much memory you expect your job will use.  mem=1g or mem=1024m
#PBS -l mem=256m

### Some portions of some clusters, such as older nodes in the PSI BATCH cluster, 
### support Myrinet GM.
### If you want to run GM programs, you must at minimum add the "gm" flag so ensure you are
### assigned nodes with Myrinet cards.
###
### gm
### gm_pci64b
### gm_pcixd
###
### For example, to run on 2 processes on each of 2 nodes with Myrinet cards you write
### #PBS -l nodes=2:ppn=2:gm

### Other options may be available too.  For example, to run on only 3GHz PSI 
### nodes, add "cpu3000" to your nodes line.
###
### Valid options are:
### PSI: cpu2333 cpu3000
### CITRIS: cpu900 cpu1300 cpu1500
#PBS -l nodes=8:ppn=2:cpu3000

### You can override the default 1 hour real-world time limit.  -l walltime=HH:MM:SS
### Jobs on the public clusters are currently limited to 10 days walltime.
#PBS -l walltime=1:00:00

### Switch to the working directory; by default Torque launches processes from your home directory.
### Jobs should only be run from /home, /project, or /work; Torque returns results via NFS.
echo Working directory is $PBS_O_WORKDIR
cd $PBS_O_WORKDIR 

### Run some informational commands.
echo Running on host `hostname`
echo Time is `date`
echo Directory is `pwd`
echo This jobs runs on the following processors:
echo `cat $PBS_NODEFILE`
### Define number of processors
NPROCS=`wc -l < $PBS_NODEFILE`
echo This job has allocated $NPROCS cpus

### Use gexec to run a program on the assigned processors.  $GEXEC_SVRS is automatically defined for you.
gexec -n 0 myprogram

### Alternatively, run a parallel MPI executable.
mpirun -v -machinefile $PBS_NODEFILE -np $NPROCS mympiprogram

### Or, if you're running on only one node, run your executable directly:
/some/path/to/a/binary