Training

Supercomputing for Social Scientists with R

Would you like to understand how to work with a supercomputer and translate your R workflow from a graphical-user-interface (GUI) on your desktop to a scripting/automated workflow that leverages the resources of a supercomputer?

Mensen op druk zebrapad
24 May 2022
Time
9.00-17.00
Location
Online

In this day long course you will learn the basics of supercomputing. The course starts with an introduction to the architecture of a supercomputer (a Linux cluster managed by a batch job scheduling system). With a hands-on session focused on connecting to, interacting with, and running jobs on a supercomputer.

The second half of the course is focused on translating your R workflow from a GUI (Rstudio) workflow on your desktop to a scripting/batch environment on the supercomputer. Topics covered here will be efficient programming, parallel computing, leveraging the SLURM job manager to submit your job/analysis to the supercomputer.

In this course you will:

  • Do practical exercises to learn how to use the national compute cluster Lisa and the national supercomputer effectively and how you can perform your tasks with minimal effort in the shortest possible time.
  • Experience how to achieve high performance with R using the supercomputing facilities at SURF

Requirements:

  • Basic knowledge of Linux (Unix shell) is useful
  • Basic knowledge of R is useful

You should have:

Your own laptop with an up-to-date browser and a terminal emulator (Putty for users of the OSSC/CBS RA environment). The use of the operating systems Linux and macOS is preferred, but not mandatory. For Windows users we recommend to download MobaXterm (portable version) as terminal emulator. If you are limited in choice Putty also works.

Acknowledgements

This course is sponsored by the ODISSEI project (with kind collaboration from Erik-Jan van Kesteren, from the University of Utrecht) and the EuroCC project.