The modules allow you to dynamically modify the environment variables needed to run your code (essentially PATH, LD_LIBRARY_PATH or even MAN_PATH), depending on the module you are loading.
They are in the form <software name>/<version>
The complete list of cluster modules can be displayed with the command:
module spider
The modules are organized hierarchically, in the form of dependencies. For example, to use PnetCDF you must first load an MPI implementation. The modules that you can load immediately are those displayed with:
module av
We are trying to create modules for you that already include dependencies. Nevertheless, to load the desired module, it is sometimes necessary to first have loaded one (or more) module (s) on which the desired module depends. To find out what these modules are, type the command:
module spider <desired module>
To consult the list of modules installed on the cluster, you can also click here.
To obtain more details about a module (which version of compiler, which environment variables are created when loading the module), call this command. Please note that you should load the module’s dependencies before using the help command.
module help <module name>
To load a module, use the command:
module load <module name>
Please note that software may have several versions installed, so it will have a module for each version. Among these modules, there is one that will be loaded by default if you do not specify the version number. So be sure to include the version number in the module name, otherwise you may not be loading the correct module.
To view which modules are currently loaded:
module list
To unload a module:
module unload <module name>
To unload your entire list of loaded modules, use:
module purge
It is for example recommended to use this command at the start of a batch, before loading your modules.