To simplify the design of ScaLAPACK, and because the BLAS have proven to be useful tools outside LAPACK, we chose to build a parallel set of BLAS, called PBLAS [26, 104], which perform message-passing and whose interface is as similar to the BLAS as possible. This decision has permitted the ScaLAPACK code to be quite similar, and sometimes nearly identical, to the analogous LAPACK code.

We hope that the PBLAS will provide a distributed memory standard , just as the BLAS have provided a shared memory standard. This would simplify and encourage the development of high performance and portable parallel numerical software, as well as providing manufacturers with a small set of routines to be optimized. Further details of the PBLAS can be found in [26], [104], and Appendix D.2.

Tue May 13 09:21:01 EDT 1997