Parallel numerical simulations of anisotropic and heterogeneous diffusion equations with GPGPU - Archive ouverte HAL
Poster De Conférence Année : 2014

Parallel numerical simulations of anisotropic and heterogeneous diffusion equations with GPGPU

Résumé

The aim of this work is to show how the anisotropic and heterogeneous diffusion equations can be solved numerically through the GPU technologies obtaining a speed-up improvement. Particularly, we want pro- vide a decision-making tool to choose the appropriate binomial algorithm-GPU memory region to get the best performance in terms of computational efficiency and reduction of calculation times. The numerical simulations have been performed using finite differences method for space discretization and Euler method for time discretization, the numerical schemes have been implemented using the NVIDIA CUDA GPU.
Fichier principal
Vignette du fichier
Pera_Simeoni_poster_2014.pdf (600.21 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02523859 , version 1 (29-03-2020)

Identifiants

  • HAL Id : hal-02523859 , version 1

Citer

Donato Pera, Chiara Simeoni. Parallel numerical simulations of anisotropic and heterogeneous diffusion equations with GPGPU. Perspectives of GPU computing in Physics and Astrophysics, Sep 2014, Rome, Italy. . ⟨hal-02523859⟩
34 Consultations
23 Téléchargements

Partager

More