Psst, remember the cheatsheet!

Supercomputer setup

Supercomputer setup#

A supercomputer has a lot of nodes which have the same components as your laptop or desktop computer: CPUs (sometimes also called processors or cores), memory (or RAM), and disk space. However, the supercomputer has some additional/specialized components:

  • When you want to execute a program on the supercomputer, it has to be boxed into an abstraction layer called job.

  • Login nodes are used to set up jobs (and to launch them)

  • Jobs are run on the compute nodes

  • A batch job system (scheduler) is used to run and manage the jobs

    • On CSC supercomputers, we use Slurm

  • CSC supercomputers use Lustre as the parallel distributed file system

  • When you login to CSC’s supercomputers, you enter one of the login nodes of the computer

Login node etiquette

Which of the following tasks would suit to run on the login node?

  1. python join_dataframes.py

  2. make

  3. create_directories.sh

  4. qgis

  5. tar -xzf mytool.tar.gz

Exploring the login node(s)

Let’s start a login shell from the Puhti webinterface. Which login node are you on?

username@puhti-loginXX

Close the shell and open another one, which login node are you on now? puhti-loginXX is also called the hostname which you can also retrieve using the hostname command.

-> Every time you login, you will get to a random login node. This usually does not matter, but has some side effects in more advanced topics, so it is good to always be aware where you are within the supercomputer.