FORGE(R) 90's Distributed Memory Parallelizer spreads loops and distributes data arrays for MIMD architectures interactively. The parallelized program is fully scalable, with calls to APR's parallel runtime library, interfacing any of the popular communication packages such as PVM, Express, and Linda, or native message passing systems.
With FORGE's SPMD (Single Program, Multiple Data) parallelization strategy, the same program runs on each processor while selected DO loops are re-written to automatically distribute their iterations across the processors.
John M. Levesque
President Applied Parallel Research, Inc. 550 Main Street Suite I Placerville, CA 95667 USA 916-621-1600 fax:916-621-0593 levesque@apri.com