Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

idPBS
Card
labelPBS Directives

The filenames, paths, email addresses, and some values below are things you will probably need to change. In some cases, values/names have been used to demonstrate possibilities that you could employ (in a slightly different way). Apart from the -l options, no option should appear on multiple lines.

Directive(s)

Description of purpose

#PBS -V

Export environment variables to the batch job

#PBS -c n
#PBS -c s
#PBS -c enabled

No checkpointing to be performed
Checkpointing is to be done on a job at pbs_mom shutdown
Checkpointing is allowed but must be explicitly invoked by a qhold or qchkpt command

#PBS -d /fast/jc123456

Defines the working directory path to be used for the job

#PBS -j oe
#PBS -o /tmp/output.$PBS_O_JOBID

Merge standard output and standard error streams into the named file

#PBS -m ae
#PBS -M john.doe@jcu.edu.au
#PBS -M joe.blogg@my.jcu.edu.au

Send mail at batch job abort/exit to the Email address provided

#PBS -N job_name

Assign a name to the batch job

#PBS -l pmem=4gb

Request that 4GB of memory be reserved for the batch job

#PBS -l nodes=1:ppn=2

Request that 2 CPU cores be reserved for the batch job

Card
labelPBS Variables

The variables listed in the table below are commonly used within a PBS script file.

Variable

Description

PBS_JOBNAME

Job name specified by the user

PBS_O_WORKDIR

Working directory from which the job was submitted

PBS_O_HOME

Home directory of user submitting the job

PBS_O_LOGNAME

Name of user submitting the job

PBS_O_SHELL

Script shell

PBS_O_JOBID

Unique PBS job id

PBS_O_HOST

Host on which job script is running

PBS_QUEUE

Name of the job queue

PBS_NODEFILE

File containing line delimited list on nodes allocated to the job

PBS_O_PATH

Path variable used to locate executables within the job script

Note: On multi-core systems, a node (line in PBS_NODEFILE) will identify the hostname and a CPU core.

...

labelSingle 1-CPU Job

This example runs PAUP on the input file input.nex that resides in the current working directory. A file (here we'll name it pbsjob) is created with the contents:

...

Important:  Most software will only consume 1 CPU core - e.g., requesting 8 CPU cores for a PAUP job blocks other people using the unused 7 CPU cores.  Example 1 below would likely be the most users should be basing their job scripts from.  If in doubt, contact HPRC staff.


For more information about PBSPro please click to see guide.  For a brief description of PBS directives provided in examples below, see the "Brief Explanation of PBS directive used in examples above" section immediately following the final example PBS script.

HPC staff should be able to assist researchers needing help with PBS scripts.


Singularity Example

The following PBS script requests 1 CPU core, 2GB of memory, and 24 hours of walltime

No Format
#!/bin/bash
#PBS -j oe
#PBS -m ae
#PBS -N JobName1
#PBS -M FIRSTNAME.LASTNAME@jcu.edu.au
#PBS -l walltime=24:00:00
#PBS -l select=1:ncpus=1:mem=2gb

cd $PBS_O_WORKDIR
shopt -s expand_aliases
source /etc/profile.d/modules.sh
echo "Job identifier is $PBS_JOBID"
echo "Working directory is $PBS_O_WORKDIR"

module load R/4.1.2
R ...    # Replace ... with your arguments & options.

Module Examples

Section


Column
width50%

Example 1:

The following PBS script requests 1 CPU core, 2GB of memory, and 24 hours of walltime for the running of "paup -n input.nex".

No Format
#!/bin/bash
#PBS -
c s #PBS -
j oe
#PBS -m ae
#PBS -N 
jobname
JobName1
#PBS -
l pmem=5gb
M FIRSTNAME.LASTNAME@jcu.edu.au
#PBS -l walltime=
500
24:00:00
#PBS -
M your.name@jcu.edu.au ncpu=`wc -l $PBS_NODEFILE | awk '{print $1}'` echo "------------------------------------------------------
l select=1:ncpus=1:mem=2gb

cd $PBS_O_WORKDIR
shopt -s expand_aliases
source /etc/profile.d/modules.sh
echo "Job identifier is $PBS_JOBID"
echo "Working 
This job is allocated "$ncpu" CPU cores on " cat $PBS_NODEFILE | uniq echo "------------------------------------------------------" echo "PBS: Submitted to $PBS_QUEUE@$PBS_O_HOST" echo "PBS: Working directory is $PBS_O_WORKDIR" echo "PBS:
directory is $PBS_O_WORKDIR"

module load paup
paup -n input.nex

If the file containing the above content has a name of JobName1.pbs, you simply execute qsub JobName1.pbs to place it into the queueing system.

Example 3:

The following PBS script requests 20 CPU cores, 60GB of memory, and 10 days of walltime for running of an MPI job.

No Format
#!/bin/bash
#PBS -j oe
#PBS -m ae
#PBS -N JobName3
#PBS -M FIRSTNAME.LASTNAME@my.jcu.edu.au
#PBS -l walltime=240:00:00
#PBS -l select=1:ncpus=20:mem=60gb

cd $PBS_O_WORKDIR
shopt -s expand_aliases
source /etc/profile.d/modules.sh
echo "Job identifier is $PBS_JOBID"
echo "
PBS: Job name
Working directory is $PBS_
JOBNAME" echo "------------------------------------------------------"
O_WORKDIR"

module load migrate
module load mpi/openmpi
mpirun -np 20 -machinefile $PBS_NODEFILE migrate-n-mpi ...

If the file containing the above content has a name of JobName3.pbs, you simply execute qsub JobName3.pbs to place it into the queueing system.


Column
width50%

Example 2:

The following PBS script requests 8 CPU cores, 32GB of memory, and 3 hours of walltime for running of 8 MATLAB jobs in parallel.

No Format
#!/bin/bash
#PBS -j oe
#PBS -m ae
#PBS -N JobName2
#PBS -M FIRSTNAME.LASTNAME@my.jcu.edu.au
#PBS -l walltime=3:00:00
#PBS -l select=1:ncpus=8:mem=32gb

cd $PBS_O_WORKDIR
shopt -s expand_aliases
source /etc/profile.d/modules.sh
echo "Job identifier is $PBS_JOBID"
echo "Working directory is $PBS_O_WORKDIR"

module load
paup paup -n input.nex

To submit the job for execution on a HPRC compute node simply enter the command:

No Format
qsub pbsjob

If you know this job will require more than 4GB but less than 8GB of RAM, you could use the command:

No Format
qsub -l nodes=1:ppn=2 pbsjob

If you know this job will require more than 8GB but less than 16GB of RAM, you could use the command:

No Format
qsub -l nodes=1:ppn=8 pbsjob

The reason for the special cases (latter two) is to guarantee memory resources for your job. If memory on a node is overallocated, swap will be used. Job(s) that are actively using swap (disk) to simulate memory could take more than 1000 times longer to finish than a job running on dedicated memory. In most cases, this will mean your job will never finish.

...

labelMultiple 1-CPU Jobs

Using Job Arrays

Users with a knowledge of shell scripting (e.g., bash) may choose to take advantage of job arrays. This feature significantly reduces load on our Torque/Maui server (compared to lots of individual job submissions). The example below (assume the file name is pbsjob), will only be useful as a guide

No Format
 matlab
matlab -r myjob1 &
matlab -r myjob2 &
matlab -r myjob3 &
matlab -r myjob4 &
matlab -r myjob5 &
matlab -r myjob6 &
matlab -r myjob7 &
matlab -r myjob8 &
wait    # Wait for background jobs to finish.

If the file containing the above content has a name of JobName2.pbs, you simply execute qsub JobName2.pbs to place it into the queueing system.

Example 4:

The following PBS script request uses job arrays. If you aren't proficient with bash scripting, using job arrays could be painful. The example below has each sub-job requesting 1 CPU core, 1 GB of memory, and 80 minutes of walltime.

No Format
#!/bin/bash
#PBS -
c s #PBS -
j oe
#PBS -m ae
#PBS -N 
jobarray
ArrayJob
#PBS -M 
your
FIRSTNAME.
name@jcu
LASTNAME@jcu.edu.au
#PBS -l 
pmem=2gb
walltime=1:20:00
#PBS -l 
walltime
select=
9
1:
00:00
ncpus=1:mem=1gb

cd $PBS_O_WORKDIR
shopt -s expand_aliases
source /etc/profile.d/modules.sh

module load matlab
matlab -r myjob$PBS_ARRAYID
No Format

Issuing the command

No Format
qsub -S /bin/bash -t 1-8 pbsjob

will see 8 jobs run under one major identifier. To view status of individual jobs in the array. The above example is identical (in terms of what jobs would be executed) to the one in the "Do It Yourself" section below.

Chances are you may need more advanced features of the scripting language than what is shown above. HPRC staff will endeavour to provide assistance with job arrays, if requested.

Do It Yourself

There are several legitimate reasons for wanting to run multiple single processor jobs in parallel within a single PBS script. For example, you may want to run 8 MATLAB jobs which require a toolbox that only has 4 licensed users. Only 1 MATLAB license is checked out if all 8 jobs are run on the same system. An example PBS script to do this task would look like

If the file containing the above content has a name of ArrayJob.pbs and you will be running 32 sub-jobs, you simply use qsub -t 1-32 ArrayJob.pbs to place it into the queueing system.

Note: I haven't done extensive testing of job arrays.


Example 5:

The following script is a rework of Example 2 to use the /fast/tmp filesystem for a hyperthetical workflow that is I/O intensive.  This example assumes 1 output file per job.

Note

Usage of /fast/tmp

Please make sure you first create an place all files in a folder that matches your jc number eg: jcXXXXXXXX


No Format
#!/bin/bash
#PBS -c s
#PBS -j oe
#PBS -m ae
#PBS -N jobnameJobName2
#PBS -M yourFIRSTNAME.LASTNAME@my.name@jcujcu.edu.au
#PBS -l walltime=10003:00:00
#PBS -l nodesselect=1:ppnncpus=8
#PBS -l pmem:mem=32gb

ncpu=`wc -lcd $PBS_O_NODEFILE`WORKDIR
echoshopt "------------------------------------------------------"
echo " This job is allocated "$ncpu" CPU cores on "
cat $PBS_NODEFILE | uniq
echo "------------------------------------------------------"
echo "PBS: Submitted to $PBS_QUEUE@$PBS_O_HOST"
echo "PBS: -s expand_aliases
source /etc/profile.d/modules.sh
echo "Job identifier is $PBS_JOBID"
echo "Working directory is $PBS_O_WORKDIR"

echo "PBS: Job identifier is $PBS_JOBID"
echo "PBS: Job name is $PBS_JOBNAME"
echo "------------------------------------------------------"

cd $PBS_O_WORKDIR
source /etc/profile.d/modules.shmkdir -p /fast/tmp/jc012345/myjobs
cp -a myjob1.m myjob2.m myjob3.m myjob4.m myjob5.m myjob6.m myjob7.m myjob8.m /fast/tmp/jc012345/myjobs/
pushd /fast/tmp/jc012345/myjobs

module load matlab
matlab -r myjob1 &
matlab -r myjob2 &
matlab -r myjob3 &
matlab -r myjob4 &
matlab -r myjob5 &
matlab -r myjob6 &
matlab -r myjob7 &
matlab -r myjob8 &
wait    # Wait for background jobs to finish.

To submit the job for execution on a HPRC compute node simply enter the command:

No Format
qsub pbsjob

Note: The echo commands in the PBS script example above are informational only.

...

labelMPI/PVM/OpenMP Jobs
No Format

#!/bin/bash
#PBS -V
#PBS -m abe
#PBS -N migrate
#PBS -l pmem=62GB
#PBS -l nodes=1:ppn=24
#PBS -l walltime=240:00:00
#PBS -M your.email@my.jcu.edu.au
cd #PBS_O_WORKDIR
module load openmpi
module load migrate
mpirun -np 24 -machinefile $PBS_NODEFILE migrate-n-mpi ...
PBS
Card
MPI/PVM/OpenMP JobsMPI/PVM/OpenMP Jobs
Deck of Cards
PBS
cp -a out1.mat out2.mat out3.mat out4.mat out5.mat out6.mat out7.mat out8.mat $PBS_O_WORKDIR/
popd
rm -rf /fast/tmp/jc012345/myjobs

Consider the possibility that you may be running more than one workflow at any given time.  Using subdirectories is a good way of segregating workflows (at a storage layer).

Brief Explanation of PBS directive used in examples above

DirectiveDescription of impact
#PBS -j oeMerge STDOUT & STDERR streams into a single file
#PBS -m aeSend an Email upon job abort/exit.
#PBS -N ...Assign a meaningful name to the job (replace ... with 1 "word" - e.g., test_job).

#PBS -M ...

Email address that PBSPro will use to provide job information (if desired)
#PBS -l walltime=HH:MM:SSAmount of clock time that your job is likely to required.
#PBS -l select=1:ncpus=X:mem=YgbRequest 1 chunk of  "X CPU cores" & "Y GB of RAM".
The "select=1:" is not really required as it is the default.
Due to JCU cluster size, requests for more than 1 chunk should/will be rejected.

Brief Details on some extra PBS/Torque directives

Directive(s)

Description of purpose

#PBS -d <PATH_TO_DIRECTORY>

Sets the working directory for you job to <PATH>.

#PBS -o <OUTPUT_FILE_PATH>

Explicit specification of file that will hold the standard output stream from you job.

#PBS -V

Export environment variables to the batch job

For full details on directives that can be used, use "man qsub" on a HPC login node or look at online documentation for Torque.

PBS/Torque Variables

The following variables can be useful within your PBS job script.  Some are present in the examples above.

Variable

Description

PBS_JOBNAME

Job name specified by the user

PBS_O_WORKDIR

Working directory from which the job was submitted

PBS_O_HOME

Home directory of user submitting the job

PBS_O_LOGNAME

Name of user submitting the job

PBS_O_SHELL

Script shell

PBS_O_JOBID

Unique PBS job id

PBS_O_HOST

Host on which job script is running

PBS_QUEUE

Name of the job queue

PBS_NODEFILE

File containing line delimited list on nodes allocated to the job (may be required for MPI jobs).

PBS_O_PATH

Path variable used to locate executables within the job script