Implementing Basic Computational Kernels of Linear Algebra on Multicore

Author

Michailidis, Panagiotis D. ; Margaritis, Konstantinos G.

Author_Institution

Dept. of Balkan Studies, Univ. of Western Macedonia, Florina, Greece

fYear

2012

fDate

5-7 Oct. 2012

Firstpage

217

Lastpage

222

Abstract

This paper implements basic computational kernels of the scientific computing such as matrix - vector product, matrix product and Gaussian elimination on multi-core platforms using several parallel programming tools. Specifically, these tools are Pthreads, OpenMP, Intel Cilk++, Intel TBB, Intel ArBB, SMPSs, SWARM and Fast Flow. The aim of this paper is to present an unified quantitative and qualitative study of these tools for parallel computation of scientific computing kernels on multicore. Finally, based on this study we conclude that the Intel ArBB and SWARM parallel programming tools are the most appropriate because these give good performance and simplicity of programming.

Keywords

application program interfaces; linear algebra; mathematics computing; matrix multiplication; multiprocessing programs; parallel programming; FastFlow; Gaussian elimination; Intel ArBB; Intel Cilk++; Intel TBB; OpenMP; Pthreads; SMPS; SWARM; basic computational kernel implementation; linear algebra; matrix-vector product; multicore platforms; parallel computation; parallel programming tools; scientific computing kernels; Kernel; Libraries; Multicore processing; Parallel programming; Program processors; Vectors; Linear Algebra; Multi-core Programming; Parallel Processing;

fLanguage

English

Publisher

ieee

Conference_Titel

Informatics (PCI), 2012 16th Panhellenic Conference on

Conference_Location

Piraeus

Print_ISBN

978-1-4673-2720-6

Type

conf

DOI

10.1109/PCi.2012.23

Filename

6377394