A parallel algorithm for four-index transformation and MP2 energy eval
uation, for distributed memory parallel (MIMD) machines is presented.
The underlying serial algorithm for the present parallel effort is the
four-index transform. The scheme works through parallelization over A
O integrals and, therefore, spreads the O(n(3)) memory requirement acr
oss the processors, reducing it to O(n(2)). In this sense, the scheme
superimposes a shared memory architecture onto the distributed memory
setup. A detailed analysis of the algorithm is presented for networks
with 4, 6, 8, 10, and 12 processors employing a smaller test case of 8
6 contractions. Model direct MP2 calculations for systems of sizes ran
ging from 160 to 238 basis functions are reported for 11- and 22-proce
ssor networks. A gain of at least 40% and above is observed for the la
rger systems. (C) 1997 by John Wiley & Sons, Inc.