A parallel multi-p method

A parallel implementation of the multi-p method is discussed, using the mas ter/slave model and the Parallel Virtual Machine (PVM) message passing libr ary. In a series of performance tests, significant speed-up was achieved in those typical cases for the p-version where there was sufficient computati onal granularity to justify use of the parallel method. These tests indicat e that the algorithms devised for load distribution and load balancing are sufficiently robust.;These tests also indicate that, even though communicat ion overhead in a network environment is relatively high, there is signific ant potential for scaling the method to larger processor ensembles. (C) 200 0 Elsevier Science Ltd. All rights reserved.