DocumentCode :
3229699
Title :
Performance evaluation of a parallel iterative method library using OpenMP
Author :
Kotakemori, Hisashi ; Hasegawa, Hidehiko ; Nishida, Akira
Author_Institution :
JST CREST
fYear :
2005
fDate :
1-1 July 2005
Lastpage :
436
Abstract :
The present paper discusses scalable implementations of sparse matrix-vector products using OpenMP to execute the iterative method on the SGI Altix3700, the IBM eServer p5 595 and the Sun SunFire15K. Three storage formats (CRS, BSR and DIA) for sparse matrices are evaluated. The present implementation provides satisfactory scalabilities. In some cases, an optimal storage format with data conversion should be used. In addition, the influence of the cache/memory bus architectures on the optimum choice of the storage format is examined
Keywords :
application program interfaces; cache storage; iterative methods; mathematics computing; parallel processing; software libraries; software performance evaluation; sparse matrices; IBM eServer p5 595; OpenMP; SGI Altix3700; Sun SunFire15K; cache/memory bus architectures; data conversion; parallel iterative method library; performance evaluation; sparse matrix-vector products; storage format; Computer architecture; Data conversion; Data structures; Iterative methods; Large-scale systems; Libraries; Memory architecture; Scalability; Sparse matrices; Sun;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
High-Performance Computing in Asia-Pacific Region, 2005. Proceedings. Eighth International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7695-2486-9
Type :
conf
DOI :
10.1109/HPCASIA.2005.74
Filename :
1592301
Link To Document :
بازگشت