Runtime system High order scheme Heterogeneous architecture Programming model Routing Absorbing Boundary Conditions Predictive Models Heterogeneous computing Memory footprint Hybrid parallelization Pénalisation GPU Numerical detection Deep learning Détection numérique Bit-flip Task-based Robust optimization Dense linear algebra Discontinuous Galerkin method Data locality Multicore architectures Conjugate Gradient method Optimal transportation Discrete velocity model Simulation Multicore architecture Exascale Multicore Processing Modeling Algorithmic NUMA Sparse direct solvers Optical mapping Fish-like swimming Pipeline Bandwidth Aetiological lithiasis Collision model Lossy compression Sparse direct solver FMM Automatic recognition Fluid mechanics Starpu-simgrid Distributed computing MPI Accelerators Asymptotic analysis Accelerator OpenMP DAG based runtime Fast multipole methods Sensitivity Fast multipole method Level-set Multicore Soft-erreur Low-rank compression Cholesky factorization Penalization Branch-cut-and-price Domain decomposition Cholesky Factorization Locality Inverse problem BGK model Uncertainty Quantification Adaptation de maillage Algorithms Robustesse Cutting and Packing Penalization method Column Generation High performance computing Gradient Conjugué Fault tolerance Robustness Placement LU factorization ALE Arbitrtary Lagrangian Eulerian Analyse numérique HPC BEM Sparse linear solver Load balancing Scheduling Parallelism Turbulence Heterogeneous architectures Integer programming Block low-rank compression Sensibilité Memory Contention Mesh adaptation AVX-512 Topology Runtime Task-based programming StarPU