Post-doctorate position (M/F) : Exascale Port of a 3D Sparse PIC Simulation Code for Plasma Modeling

id="offer-description" class="ecl-u-type-heading-2 ecl-u-mt-none ecl-u-mb-l">Offer Description The person recruited will be involved in the ANR project called MATURATION ( https://www.math.univ-toulouse.fr/~narski/MATURATION/ ) based on the use of sparse grids coupled with the Particle-In-Cell (PIC) approach to study plasma physics. The aim will be to port to exascale architectures an initial 3D simulation code developed as part of previous work [1]. This work will initially focus on scaling up (distributed memory), optimizing CPU algorithms (vectorization) and further GPU porting. The exploration of C++ programming models for performance portability, such as Kokkos or StarPU, will form a second part. A comparative study will evaluate the different implementations in terms of both performance and portability. Performance studies will be carried out on various CPU and GPU technologies, using local resources and access to national supercomputers. The person recruited will also be involved in setting up and sharing best practices in software engineering. This work will be carried out in close interaction with the other MATURATION project partners. [1] - Guillet, C. (2023). Approche sur grilles parcimonieuses pour accélérer la méthode PIC (PhD, Toulouse 3). - Participate in the Exascale porting of 3D sparse-PIC code, including distributed parallelism, vectorization, and GPU porting. - Investigate new programming models to achieve performance portability. - Analyze code performanc