Design patterns for sparse-matrix computations on hybrid CPU/GPU platforms

Article type: Research Article

Authors: Cardellini, Valeria | Filippone, Salvatore^; | Rouson, Damian W.I.

Affiliations: Università di Roma “Tor Vergata”, Roma, Italy. E-mails: [email protected], [email protected] | Stanford University, Stanford, CA, USA. E-mail: [email protected]

Note: [] Corresponding author: Salvatore Filippone, Dipartimento di Ingegneria Industriale, Università di Roma “Tor Vergata”, Viale del Politecnico 1, 00133 Roma, Italy. Tel.: +39 06 72597558; Fax: +39 06 72597158; E-mail: [email protected]

Abstract: We apply object-oriented software design patterns to develop code for scientific software involving sparse matrices. Design patterns arise when multiple independent developments produce similar designs which converge onto a generic solution. We demonstrate how to use design patterns to implement an interface for sparse matrix computations on NVIDIA GPUs starting from PSBLAS, an existing sparse matrix library, and from existing sets of GPU kernels for sparse matrices. We also compare the throughput of the PSBLAS sparse matrix–vector multiplication on two platforms exploiting the GPU with that obtained by a CPU-only PSBLAS implementation. Our experiments exhibit encouraging results regarding the comparison between CPU and GPU executions in double precision, obtaining a speedup of up to 35.35 on NVIDIA GTX 285 with respect to AMD Athlon 7750, and up to 10.15 on NVIDIA Tesla C2050 with respect to Intel Xeon X5650.

Keywords: Design patterns, sparse matrices, GPGPU computing

DOI: 10.3233/SPR-130363

Journal: Scientific Programming, vol. 22, no. 1, pp. 1-19, 2014