Performance evaluation of a parallel iterative method library using OpenMP
Performance evaluation of a parallel iterative method library using OpenMP
复制标题
DOI：
10.1109/hpcasia.2005.74
复制
发表时间：
2005-11
期刊：
影响因子：
--
通讯作者：
Hisashi Kotakemori;H. Hasegawa;A. Nishida
中科院分区：
文献类型：
--
作者：
Hisashi Kotakemori;H. Hasegawa;A. Nishida
本文讨论了稀疏矩阵向量乘积的可扩展实现，使用OpenMP执行迭代方法上的SGI Altix 3700，IBM eServer p5 595和Sun SunFire 15 K。对稀疏矩阵的三种存储格式（CRS、BSR和DIA）进行了评价。本实施方式提供了令人满意的可扩展性。在某些情况下，应使用具有数据转换的最佳存储格式。此外，还研究了该高速缓存/存储器总线结构对存储格式最佳选择的影响
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