Educators & Organizers
Teresa Müller (RBC), Silvia Di Giorgio (ZB MED – Associated Member), Helena Vela (HPCNow! - Do IT Now Group), Alan O'Cais (University of Barcelona)
Date:
October 21, 2025 - October 23, 2025
From 09:00 to 16:00 (CEST)
Location:
Online
Contents:
Here, we are offering a 3-day workshop, composed of 3 full-day sessions, with the primary goal of introducing participants to the daily tasks of HPC system administrators through realistic scenarios using industry-standard tools and technologies. This practical training is intended for junior system administrators, technical staff, or Linux users transitioning into HPC environments. The program blends foundational system administration concepts with hands-on HPC-specific practices.
Day 1 focuses on system administration: user and group management, permissions, filesystems, package and service management, and firewalls.
Day 2 introduces HPC cluster-specific operations, with an in-depth look at the Slurm workload manager, and modern container technologies like Docker and Singularity.
Day 3 covers automation (Ansible), monitoring (Prometheus, Grafana), and software stack management (EasyBuild, EESSI, and Spack).
This workshop offers a comprehensive, practical introduction to HPC system administration, empowering junior and aspiring administrators to confidently support and grow HPC infrastructures.
Learning goals:
By the end of this workshop, you will be able to:
- Apply core Linux system administration skills in an HPC context
- Manage shared filesystems, users, services, and packages
- Set up and administer Slurm workload manager from a system perspective
- Deploy and support scientific applications in containers
- Automate configuration using Ansible
- Monitor and troubleshoot cluster health with Prometheus and Grafana
- Build and manage HPC software environments with EasyBuild and EESSI
Prerequisites:
- The lessons require you to have access to a terminal application with ssh capabilities. If it is unclear what this requirement means, please click here for guidance on how to make this available for your operating system
- There is no need for programming or informatics skills but a prior knowledge of file systems and the Unix shell is required. If you wish to participate but do not meet these prerequisites, we recommend watching the video recording from our previous workshop, available in the BioNT Lhumos space here.
- PC/Laptop with an up-to-date browser. Chrome, Safari and Firefox browsers are all supported (some older browsers, including Internet Explorer version 9 and below, may not be)
Keywords:
HPC
Tools:
Ansible, Prometheus, Grafana, EasyBuild, EESSI, Spack,
Contact:
Registration:
https://www.cecam.org/workshop-details/system-administration-for-hpc-1465