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