PERFORMANCE STUDY OF THE DOMAIN DECOMPOSITION METHOD WITH DIRECT EQUATION SOLVER FOR PARALLEL FINITE-ELEMENT ANALYSIS

A parallel performance of the domain decomposition method with direct LDU algorithm of condensation and solution is studied. Typical subdoma ins arising after division of a square domain are considered, and oper ation count equations for all steps of the numerical procedure are der ived. The parallel efficiency model is developed using operation count equations and message passing estimates. It is shown how to achieve i nterprocessor load balancing by partitioning a domain into unequal sub domains. The evaluation of the parallel efficiency model and performan ce studies for a square finite element domain are performed on the IBM SP2 computer with 4, 6 and 8 processor nodes. It is found that proper load balancing of the domain decomposition algorithm with direct solu tion of equation systems provides acceptable parallel efficiency for m ultiprocessor computers: 95% for the 6-processor configuration and 85% for the 8-processor configuration.