TOWARD HIGH-PERFORMANCE COMPUTATIONAL CHEMISTRY .2. A SCALABLE SELF-CONSISTENT-FIELD PROGRAM

We discuss issues in developing scalable parallel algorithms and focus on the distribution, as opposed to the replication, of key data struc tures. Replication of large data structures limits the maximum calcula tion size by imposing a low ratio of processors to memory. Only applic ations which distribute both data and computation across processors ar e truly scalable. The use of shared data structures that may be indepe ndently accessed by each process even in a distributed memory environm ent greatly simplifies development and provides a significant performa nce enhancement. We describe tools we have developed to support this p rogramming paradigm. These tools are used to develop a highly efficien t and scalable algorithm to perform self-consistent field calculations on molecular systems. A simple and classical strip-mining algorithm s uffices to achieve an efficient and scalable Fock matrix construction in which all matrices are fully distributed. By strip mining over atom s, we also exploit all available sparsity and pave the way to adopting more sophisticated methods for summation of the Coulomb and exchange interactions. (C) 1996 by John Wiley & Sons, Inc.