ocelote_cpu_array.pbs

#!/bin/bash

### This will be the researcher who supported your use of the HPC.
#PBS -W group_list=[your researcher here]

### Researchers can buy in for priority queueing. However, the amount of time 
### they can use for this is limited.  Everyone has access to unlimited
### windfall.  However, any priority jobs will go first.  The queues on Ocelote
### are windfall, standard, qualified, high, ssp_standard
#PBS -q windfall

### Most Nodes have 28 cores on Ocelote.  This configuration shows a request
### for 1 node with 1 core.  If you have a job that needs more than 28 cores,
### you can increase the number of nodes but be careful because that MULTIPLIES
### the number of requested cores.  Requesting 2 nodes with 15 cores will give
### you 30 total cores, for example.  mem is the total amount of memory (per
### node, 168gb max) available for the job.  On Ocelote, this is the total
### number of cores multiplied by 6gb.  pcmem is the amount of memory per core.
### This is the same as the limit of 6 per core.  The more efficiently you
### request your resources, the less time your job will spend in the queue and
### the more resources will be available for others to use.  If you're not sure
### how much you should request, please contact the consult team.

#PBS -l select=1:ncpus=1:mem=6gb:pcmem=6gb

### This is the amount of time you think your job will take to run.
### Not sure how long it will take?  Request the max of 10 days.
### (240:00:00)                  #PBS -l walltime=240:00:00
### This request shows 5 minutes #PBS -l walltime=00:05:00
#PBS -l walltime=[your walltime here]

### An array of 5 jobs.  Each job in the array will have a request for
### resources as shown above.
#PBS -J 1-5

### This will show which node your job is running on.
echo 'This job is running on:'
hostname

### use PBS script to display index
echo job array index $PBS_ARRAY_INDEX

### If you have any trouble or questions, please contact:
### hpc-consult@list.arizona.edu