First Hands-on Experience on a HPC Cluster
Course further information
In this workshop, you will learn how to use high-performance computing (HPC) clusters for your scientific work.
While HPC clusters are composed of components similar to those found in PCs or workstations, they are used in a very different way. This is mainly owed to the fact that they consist of many computers networked together and that they are shared by multiple users.
We start this workshop with explaining to you the inner structure of a typical HPC cluster and highlight the differences to a workstation. You will then learn how to use the Slurm workload manager, which is used on the university cluster “Draco” to distribute compute jobs across the hardware. We will also explore the various types of batch jobs and interactive tasks. During the hands-on sessions, you will submit your first compute jobs to the cluster and hopefully enjoy their results. Finally, we will provide guidance on how to install and use your own parallel software.
This workshop is held in person, online particiaption is not possible. The course language is English.
Requirements
To participate, you need a user account of the Univeristy of Jena, which needs to be entered during registration. You should also have basic familiarity with Linux and using the command or the curiosity to explore it.
You may use your own laptop or one of the pool PCs.
