The CSF2 has been replaced by the CSF3 - please use that system! This documentation may be out of date. Please read the CSF3 documentation instead. To display this old CSF2 page click here. |
IDL
Overview
IDL provides a dynamically typed visualization programming language and tools for data viz.
Version 8.6 is installed on the CSF.
Restrictions on use
IDL is licenced software. To use this software you will need to show you have a licence and be added to the idl
Unix group. It is currently restricted to a specific research group.
Any permitted usage must adhere to the following license conditions:
IDL licences supplied under this agreement are solely used by researchers in astronomy and solar systems science (including Solar Terrestrial Physics) within the STFC remit.
Please contact its-ri-team@manchester.ac.uk for more information about this software.
Set up procedure
You can add the IDL software binaries to your $PATH by running
module load apps/binapps/idl/8.6
This will also set up the required licence paths.
Running the application
The application is launched by issuing the command idl
.
If you need to run the software for more than a few minutes using significant CPU or memory resources you should run it via a compute node.
Serial batch job
Make sure you have loaded the modulefile. Then create a jobscript similar to the following:
#!/bin/bash #$ -cwd #$ -V # -- the commands to be executed (programs to be run) : idl < test.pro # # Change to be the name of your IDL input file
Submit the job using
qsub myjobscript
where myjobscript
is the name of your jobscript created above.
Example of an interactive IDL session
Use the qrsh
command to request interactive compute resources:
qrsh -l inter -l short
#
# Wait until you have been logged in to a compute node. Then:
# Set up to run IDL
module load apps/binapps/idl/8.6
# Run idle interactive (will give you the IDL
prompt)
idl
# Alternatively, supply an IDL input file
idl < myscript.pro
Please also consult the qrsh documentation for more detailed instructions on interactive use of compute nodes.
Parallel Job Submission
IDL should only be run on a single node (shared memory parallelism)
Instructions coming soon.
Further info
Examples and help are available in $IDL_DIR/examples
and $IDL_DIR/help