Fast and green computing with graphics processing units for solving sparse linear systems - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

Fast and green computing with graphics processing units for solving sparse linear systems

Abal-Kassim Cheik Ahamed
A. Desmaison
  • Fonction : Auteur
F. Magoules

Résumé

In this paper, we aim to introduce a new perspective when comparing highly parallelized algorithms on GPU: the energy consumption of the GPU. We give an analysis of the performance of linear algebra operations, including addition of vectors, element-wise product, dot product and sparse matrix-vector product, in order to validate our experimental protocol. We also analyze their uses within conjugate gradient method for solving the gravity equations on Graphics Processing Unit (GPU). Cusp library is considered and compared to our own implementation with a set of real matrices arrising from the Chicxulub crater and obtained by the finite element discretization of the gravity equations. The experiments demonstrate the performance and robustness of our implementation in terms of energy efficiency.

Dates et versions

hal-01273943 , version 1 (15-02-2016)

Identifiants

Citer

Abal-Kassim Cheik Ahamed, A. Desmaison, F. Magoules. Fast and green computing with graphics processing units for solving sparse linear systems. 16th International Conference on High Performance and Communications (HPCC 2014), Aug 2014, Paris, France. ⟨10.1109/hpcc.2014.25⟩. ⟨hal-01273943⟩
71 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More