New partitioning schemes for parallel modified Gram-Schmidt orthogonalization

被引：2

作者：

Oliveira, S ^{[1
]}

Soma, T ^{[1
]}

机构：

[1] Texas A&M Univ, Dept Comp Sci, College Stn, TX 77843 USA

来源：

THIRD INTERNATIONAL SYMPOSIUM ON PARALLEL ARCHITECTURES, ALGORITHMS, AND NETWORKS, PROCEEDINGS (I-SPAN '97) | 1997年

关键词：

Gram-Schmidt algorithms; orthogonalization; MPI; data partitioning; matrix computations;

D O I：

10.1109/ISPAN.1997.645102

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

In this paper we are concerned with parallel implementation of row-oriented Gram-Schmidt orthogonalization. For the data partitioning four types of columnwise partitioning schemes were considered: column. (1-col), block, cyclic and block-cyclic (b-c) partitioning. Analytical models for parallel execution time required by these implementations are derived and compared with numerical results. The best partitioning scheme is shown theoretically and by numerical results.

引用

页码：233 / 239

页数：7