CASTIEL2: Multi-GPU AI Train the Trainer Workshop

Join us for an exclusive event hosted by HLRS, in collaboration with CASTIEL2 and EuroCC2, Europe's HPC ecosystem. The Train the Trainer Program is provided in conjunction with the regular AI and multi-GPU courses. This Train-the-Trainer program runs alongside the standard AI and multi-GPU computing courses. While the regular course focuses on practical skills for building and scaling AI workloads across multiple GPUs, the Train-the-Trainer track is dedicated to preparing future instructors to teach these topics effectively.

Too few people are currently able to offer advanced AI and multi-GPU training at the level required by scientists, engineers, and PhD students who need to scale AI models efficiently or optimize distributed training workflows. Across Europe, only a limited number of HPC and AI centres provide such specialized courses. With this program, we aim to empower additional trainers and centres to deliver high-quality AI and multi-GPU training - either across Europe or within their own countries.

During this training event, participants will gain a deep understanding of modern multi-GPU and distributed AI technologies, along with practical techniques for teaching these concepts effectively. Expert instructors will guide attendees through state-of-the-art tools, frameworks, and best practices for scaling AI workloads, while also providing pedagogical strategies, teaching resources, and hands-on experience designed to prepare participants to deliver their own AI and multi-GPU training courses.

The training event is co-organised by HLRS, JSC, SURF, CSC, VSC, Cyfronet AGH, RISE, LiU, for EuroCC Poland, NCC Netherlands, NCC Hungary, NCC Belgium, NCC Italy, NCC Finland, SiDE and EuroCC Sweden, all National Competence Centres for High-Performance Computing, CASTIEL2 and EuroCC2, and is designed for everyone interested in the NCCs and the CoEs.

Veranstaltungsort

Online course
Organizer: HLRS, University of Stuttgart, Germany

Veranstaltungsbeginn

02. Feb. 2026
09:00

Verstaltungsende

05. Feb. 2026
17:00

Sprache

Englisch

Einstiegslevel

Fortgeschritten

Themenbereiche

Daten in HPC / Deep Learning / Maschinelles Lernen

Themen

Künstliche Intelligenz

Big Data

Deep Learning

Maschinelles Lernen

Scientific Machine Learning

Zurück zur Liste

Prerequisites and content levels

You are familiar with AI frameworks.
Your centre is affiliated to one National Competence Centre (NCC) of EuroCC or CoEs.

Instructors

Main lecturers:
Ashwin Mohanan (RISE)
Mats Sjoberg (CSC)
Caspar van Leeuwen (SURF)
Gyula Ujlaki (NCC Hungary)
Eleonora Bergamaschi (CINECA)
Michele Visciarelli (CINECA)
Prof. dr. Geert Jan Bex (UHASSELT)

Agenda

- PRELIMINARY -

1st day – Monday, 4 February 2026

Morning session: Intro to the HPC system. Setup and accessing Jupyter with GPU
Afternoon session: Introduction to deep-learning

2nd day – Tuesday, 5 February 2026

Morning session: Pytorch DDP
Afternoon session: Model parallelism with Pytorch

3rd day – Wednesday, 6 February 2026

Morning session: Pytorch Lightning
Afternoon session: LLM, Finetuning. HuggingFace Accelerate, Deepspeed

4rd day – Thursday, 7 February 2026

Morning session: Ray + Retrieval Augmented Generation
Afternoon session: Hyperparameter tuning

Registration

For lists of EU and EuroCC countries have a look at the Horizon Europe and EuroCC website.
For list of Centres of Excellence (CoEs) have a look at the CoEs.

Only participants from EuroCC/CoE/CASTIEL can take part in this course.

Contact

training@hlrs.de

HLRS Training Collaborations in HPC and AI

HLRS is part of the Gauss Centre for Supercomputing (GCS), together with JSC in Jülich and LRZ in Garching near Munich. SIDE is the German National Competence Centre (NCC) for High-Performance Computing. HLRS is also a member of the Baden-Württemberg initiative bwHPC.
Since 2025, HLRS has been coordinating one of the AI Factories of the EuroHPC JU: HammerHAI.

Acknowledgements

Within the scope of the EuroCC2 project this training is offered free of charge. By participating in a training free of charge, companies receive state-aid corresponding to the regular market-prices listed in the service portfolio. Please note state-aid regulations in Germany.

EuroCC 2 and EuroCC4SEE have received funding from the European High-Performance Computing Joint Undertaking (JU) under grant agreement No 101101903 and No 101191697. The JU receives support from the European Union’s Digital Europe Programme and Germany, Bulgaria, Austria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, Greece, Hungary, Ireland, Italy, Lithuania, Latvia, Poland, Portugal, Romania, Slovenia, Spain, Sweden, France, Netherlands, Belgium, Luxembourg, Slovakia, Norway, Türkiye, Republic of North Macedonia, Iceland, Montenegro, Serbia, Bosnia and Herzegovina.

Further courses and training team

See the training overview and the Supercomputing Academy pages.
See also information about the HLRS training department and staff.

Zurück zur Liste