or i o
o i sio sic i r r o ic io
ro r s ?
a . arra a r . a
r a ar
In this paper, we describe extensions to a proposed set of linear algebra communication routines for communicating and manipulating data structures that are distributed among the memories of a distributed memory MIMD computer. In particular, recent experience shows that higher performance can be attained on such architectures when parallel dense matrix algorithms utilize a data distribution that views the computational nodes as a logical two dimensional mesh. The motivation for the BLACS continues to be to increase portability, efficiency and modularity at a high level. The audience of the BLACS are mathematical software experts and people with large scale scientific computation to perform. A systematic effort must be made to achieve a de facto standard for the BLACS.
This report outlines a second attempt to define a proposed set of linear algebra communication routines for the specification and manipulation of data structures that arise when linear algebra algorithms are implemented on distributed memory multicomputers. The scope of this set of routines is intentionally limited. We do not view this package as a complete communication library for all applications. It is intended primarily for software developers and to a lesser extent for experienced applications programmers in the area of numerical linear algebra. We see these routines complementing the existing Level 1, 2, and 3 BLAS; providing tools for the implementation of numerical algorithms in linear algebra for distributed memory MIMD machines.
It is important to realize what kind of algorithms inspired the BLACS as they stand. They are being developed as part of our effort to implement a subset of the LAPACK library on