System administration for HPC workshop - BioNT

This three-day workshop, System administration for HPC, aims to provide participants with a practical introduction to the daily tasks of high-performance computing (HPC) and system administrators. Its main objective is to familiarise junior and aspiring system administrators with industry-standard tools and practices for managing HPC environments, while also providing a foundation for technical staff or advanced Linux users transitioning into HPC cluster support.

Target audience

The target audience includes junior system administrators, IT support staff, Linux users who wish to scale up their expertise, and professionals in academia, research, or industry who need to deploy or manage scientific applications in multi-user computing environments. .

Course overview

Over the course of three full-day sessions, participants will gain skills in Linux system administration, including user and group management, permissions, filesystems, package and service management, and firewalls. They will then move on to cluster-specific operations such as managing the Slurm workload manager, and using container technologies like Docker and Singularity. The final day introduces automation with Ansible, monitoring with Prometheus and Grafana, and HPC software stack management using EasyBuild, EESSI, and Spack.

Learning outcomes

By the end of the workshop, participants will be able to confidently perform system administration tasks in an HPC context, set up and administer Slurm, deploy scientific applications in containers, automate configuration, monitor cluster health, and manage HPC software environments. This training equips attendees with practical, immediately applicable skills that can support and scale HPC infrastructures.

Registration is open until 13 October 2025, and the workshop will take place online from 21 to 23 October 2025, running daily from 09:00 to 16:00 CEST. The training is free of charge