Cyclic block-algorithms for solving triangular systems on distributed-memory multiprocessors with mesh topology

被引:0
|
作者
Fiebach, P
机构
[1] Bergische Univ. GH Wuppertal, Fachbereich Mathematik, D-42097 Wuppertal
关键词
triangular solver; parallel block-algorithms; mesh topology; BLAS3;
D O I
10.1016/0167-8191(96)00073-7
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Parallel blocked versions of short-cut algorithms are presented for solving triangular systems with, possibly, multiple right hand sides. They mainly use BLAS3-routines for their implementation and require the data to be distributed through a square-block torus-wrap mapping. Numerical experiments on an INTEL Paragon XP/S show an efficiency of 50-75% for a wide range of block sizes and mesh forms.
引用
收藏
页码:383 / 393
页数:11
相关论文
共 26 条