This article gives an introduction to GPU usage for High Performance Computing. After setting the context, we will describe the hardware and the programming languages currently available to programmers. From these explanations we will touch on the implications of these technologies for simulation codes and try to give trends for the future.
Cet article est une introduction au calcul sur GPU pour le monde du Calcul Hautes Performances. Après avoir resitué le contexte, nous décrirons les matériels et les logiciels à la disposition des programmeurs. Sur la base de nos explications, nous esquisserons les implications de ces technologies sur les codes de calcul. Enfin, nous essayerons de dégager des tendances pour le futur.
Mots-clés : Informatique, GPGPU, Matériel, Logiciels
Guillaume Colin de Verdière 1
@article{CRMECA_2011__339_2-3_78_0, author = {Guillaume Colin de Verdi\`ere}, title = {Introduction to {GPGPU,} a hardware and software background}, journal = {Comptes Rendus. M\'ecanique}, pages = {78--89}, publisher = {Elsevier}, volume = {339}, number = {2-3}, year = {2011}, doi = {10.1016/j.crme.2010.11.003}, language = {en}, }
Guillaume Colin de Verdière. Introduction to GPGPU, a hardware and software background. Comptes Rendus. Mécanique, High Performance Computing, Volume 339 (2011) no. 2-3, pp. 78-89. doi : 10.1016/j.crme.2010.11.003. https://comptes-rendus.academie-sciences.fr/mecanique/articles/10.1016/j.crme.2010.11.003/
[1] Gene Amdahl, Validity of the single processor approach to achieving large-scale computing capabilities, in: AFIPS Conference Proceedings, 1967.
[2] Nodal discontinuous Galerkin methods on graphics processors, Journal of Computational Physics, Volume 228 ( November 2009 ) no. 21, pp. 7863-7882
Cited by Sources:
Comments - Policy