A SURVEY OF PARALLEL DIRECT-METHODS FOR BLOCK BIDIAGONAL LINEAR-SYSTEMS ON DISTRIBUTED-MEMORY COMPUTERS

Four parallel algorithms for the solution of block bidiagonal linear s ystems on distributed memory computers are presented. All the algorith ms belong to the class of direct methods. The first is a variant of th e sequential algorithm and is suitable for a small number of processor s. The remaining three algorithms are based on the parallel methods fo r banded systems and are much better suited for parallel computations on multiple processors. The arithmetical complexity functions of the p roposed algorithms are derived. The results of experiments with the fo ur algorithms implemented in Parallel Fortran on a linear array of 32 Transputers are presented and discussed.