Agave's Software Modules

Supercomputing software environments are highly domain dependent, and to prevent conflicts while balancing ease of use, the software that your research needs will likely need to be loaded into the environment as a software module.

As of July 21, 2022, there were 1,380 software modules on Agave (https://links.asu.edu/modules)! New packages are added regularly, typically at the request of researchers.

This page documents how to use the module system as to configure Agave’s supercomputing environment to properly execute your computational research workflows.

1 Listing available modules
2 Loading a module
3 Full versions are required
4 Listing Loaded Modules
5 Purging Loaded Modules
6 Using Modules in SBATCH scripts

Modules that install into/through python or anaconda are not managed through the software modules. To see if we have a conda environment for a python program load an anaconda module and list conda environments e.g.

module load anaconda/py3
conda info --envs

You can also create and manage your own conda environments. For more information see Using anaconda environments

Listing available modules

The following command will list the software modules that are available on Agave:

module -l avail

Agave’s modules may also be perused online here: https://links.asu.edu/modules

Loading a module

The following command will load the module for gcc/4.9.2

module load gcc/4.9.2

Full versions are required

In the past, users could load a module using a short name, without a version number, as follows:

module load gcc # THIS WILL NO LONGER WORK

Previously, this would load whichever module was the default for GCC.

This will no longer work because there is no default version for any module.

The full name of a module including its version must be used when loading modules.

Listing Loaded Modules

The following command will list the modules that are currently loaded:

Purging Loaded Modules

To clear out the modules that are currently loaded, perhaps because you want to load others, use the following command:

Using Modules in SBATCH scripts

Scheduled computational job scripts must specify the software environment that they need in order to execute properly. This increases the portability and consistency of computational research workflows (by providing effective documentation and definition of the necessary software environment to have the work run successfully).

On Agave, software modules are loaded with the module load <software>/<version> command pattern as shown above.

An example of this can be found here: Creating SBATCH scripts

Recall that on Agave: computational work must be scheduled through the use of batch scripts that are typically passed to the shell command, sbatch for scheduling.