🚀 Container Technology for HPC
Dive into the heart of a software revolution essential for research and industry: containerization applied to high-performance computing (HPC). This 3-hour hands-on workshop will allow you not only to understand the principles but also to create, run, and orchestrate containers tailored for HPC environments — including GPU usage and multi-node parallel computing with MPI.
Session content
🔍 Introduction: concepts and benefits of containerization for HPC
🛠 Hands-on — basic commands: getting started with essential tools
🧩 Hands-on — building your container: constructing and testing a reproducible image
⚡ Hands-on — Containers & HPC:
-
Running a container with GPU access
-
Running multiple containers in parallel across nodes with MPI
Why participate?
🎯 Highly practical workshop — you will leave with operational skills ready to use immediately.
🚀 In-demand skills: containers + GPU + MPI = key expertise for high-performance computing projects, data science, and AI.
🤝 On-site advantage: being present maximizes guidance and technical exchange. Online participation is available if you cannot attend in person.
⚠️ Important: mastery of prerequisites is necessary to fully benefit from the workshop.
Before attending — recommended checklist
-
Have basic Linux and SSH skills (connecting to a supercomputer).
-
Bring a laptop with a working SSH client.
-
(If possible) generate a public SSH key for easier access.
-
Have a ROMEO account if you plan to use ROMEO during the session (useful but not mandatory).
Date: November 10 — 2:00 PM to 5:00 PM
Mode: Hybrid (on-site attendance preferred at URCA — remote participation possible)
Prerequisites (mandatory): familiar with basic Linux commands, able to use SSH and connect to a supercomputer
Location: URCA, Moulin de la Housse Campus — Building 7, Room TD701
r
Register now — limited spots
Free training offered by CC-FR, URCA, ROMEO, and MesoNet.
