About these tutorials

Video

Watch this in our courses: 2021 January.

The “Shell crash course” is a useful prerequisite. You can watch a short version (20 min) or longer version (1 hour).

Or see the full playlist (2022 winter) (2021 summer, 2021 winter.)

Welcome to the Aalto Scientific Computing High-performance computing (HPC) tutorials. These tutorials will get you started with the Triton cluster.

Despite the HPC in the name, most of these tutorials are not about the high-performance part: instead, we get you started using and submitting jobs to the cluster. These days, many people use a cluster for simple jobs: getting more stuff done at once, not a few very big tasks. Doing the big tasks are a more specialized topic, which this will introduce you to and you will be able to use other software for that. Programming your own HPC software is out of our scope.

Not at Aalto?

These tutorials use Aalto’s cluster as an example, but they are designed to be useful to a wide audience: most clusters operate on the same principles with local configuration or practices needed. This course/these tutorials, along with a quick reference similar to ours, will be a great start to your career. (People running a cluster can check out our hint sheet to see what differences you may need to explain.)

We will point out things that may be different, but you need to consult your own reference to see how to do it:

  • The way you connect to the cluster, including remote access methods.

  • Exact names of batch partitions.

  • The slurm utility probably isn’t installed, seff may not be there.

  • Module names for software.

  • You probably don’t have our Singularity container stuff installed.

  • Parallel and GPU stuff is probably different.

What’s next?

Introduce yourself to the cluster resources at Aalto.