The first workshop of the NumPEx Accelerator working group

On June 12-13^th 2024, the Accelerator working group held the workshop “Programmation GPU” to take a first review of the current situation.

This workshop was the perfect occasion to have a comprehensive overview of the various approaches currently available for an effective use of GPUs, including direct programming, libraries, frameworks, and task-based methods.
The workshop enabled participants to leave with a clear understanding of the advantages and disadvantages of each approach and to benefit from insights and experiences with different codes across these approaches.

You will find below all the presentation materials and video recordings of the day’s events, which were held in French.

• Introduction et vue d’ensemble des journées
• Introduction aux GPUs, à la programmation CUDA

Both presented by Samuel Thibault, professor at Université de Bordeaux

• Approche framework: Arcane, API accélérateur
  Gilles Grospellier, CEA researcher
• Approche bibliothèque: GPU Programming through external scientific libraries
  Florent Pruvost, Inria researcher
• Approche langage: Kokkos / OpenMP
  Julien Bigot, CEA researcher
• Approche tâches, StarPU
  Samuel Thibault, professor at Université de Bordeaux

• Retex – Approche tâches pour l’algèbre linéaire GPU + distribué
  Antoine Jego, professor at Sorbonne Université
• Retex – Approche tâches pour l’algèbre linéaire GPU + I/O, out-of-core, composition (Chameleon)
  Florent Pruvost, Inria researcher
• Retex – Approche OpenACC: YALES2
  Vincent Moureau, CNRS researcher
• Retex – Approche OpenACC: Porting a legacy Fortran CFD HPC code to (AVBP)
  Joeffrey Legaux, engineer at CERFACS
• Retex – Approche Kokkos: Dyablo, A new hardware-agnostic AMR code for Exascale using Kokkos
  Arnaud Durocher, CEA researcher and engineer
• Retex – Rust et OpenCL pour le portage GPU (minicl)
  Philippe Helluy, professor at Université de Strasbourg

• Contributions prévues au sein du PEPR NumPEx
  Samuel Thibault, professor at Université de Bordeaux
• Brainstorming on the GPU call for proposals