Introduction to the cluster

10 October 2024, Wes Hinsley

Get started using the DIDE cluster using hipercow. We will walk through the basic idea of high performance computing, and get you running some simple jobs that will scale work off your laptop and onto our cluster.

Email as sent with instructions and context below:

Our introduction to the cluster from R (aka “hipercow”) training is this Thursday afternoon in the Board Room, 310 on the 3rd floor of SPH.

This is aimed at first-time users of the cluster, or people who have not used it before, though everyone is welcome. We assume no real prior knowledge except:

• You use R for some work that takes longer on your laptop than you would like
• You would like to run it faster
• You have time to read through some instructions on getting your passwords and directories in order - https://docs.google.com/presentation/d/13uHkcO6j2wClNJYHuUe9PS3FcVtaMZdVCC0MKS3Wbko/edit?usp=sharing

We will cover some basic ideas in parallel computing, and guide you through using the packages. By the end of the session you will have run some simple jobs on the cluster and will be able to use it in your own work.

In early 2025 we plan on running a slightly more advanced session along these lines that will focus on running many jobs at once or on jobs that are themselves parallel.

If you don’t use R but are interested in using the cluster you are very welcome, but please install R anyway to play along.

If you’re coming on Thursday afternoon, please go through the preflight instructions here first: https://docs.google.com/presentation/d/13uHkcO6j2wClNJYHuUe9PS3FcVtaMZdVCC0MKS3Wbko/edit?usp=sharing which will ensure you’ve got your passwords and folders working first.

See you there! Wes

Back to list of sessions