Applications: General info¶
Intro tutorial: Applications (this is assumed knowledge for all software instructions)
When you need software, check the following for instructions (roughly in this order):
Search the SciComp site using the search function.
module availto see if something is available but undocumented.
The issue tracker for other people who have asked - some instructions only live there.
If you have difficulty, it’s usually a good idea to search the issue tracker anyway, in order to learn from the experience of others.
See Software modules. Modules are the standard way of loading software.
See Singularity Containers. Singularity are software containers that provide an operating system within an operating system. Software will tell you if you need to use it via Singularity.
Software installation and policy¶
We want to support all software, but unfortunately time is limited. In the chart below, we have these categories (which don’t really mean anything, but in the future should help us be more transparent about what we are able to support):
A: Full support and documentation, should always work
B: We install and provide best-effort documentation, but may be out of date.
C: Basic info, no guarantees
If you know some application which is missing from this list but is
widely in use (anyone else than you is using it) it would make sense
/share/apps/ directory and create a module file. Send
your request to the tracker. We want to support as much software as
possible, but unfortunately we don’t have the resources to do
Software is generally easy to install if it is in Spack (check that package list page), a scientific software management and building system. If it has easy-to-install Ubuntu packages, it will be easy to do via singularity.
Software documentation pages¶
- COMSOL Multiphysics
- Deep learning software
- Gurobi Optimizer
- Intel Compilers
- Using Mathematica on Triton
- NVIDIA’s singularity containers
- Python environments with Conda
- Siesta & Transiesta
- Your own notebooks via