Comparative Performance of Frontal (Direct) and PCG (Iterative) Solver Based Parallel Computations of Finite Element Analysis

PARAM-10000 is latest of the series of High Performance Parallel Super Computers developed by India. It employs a distributed memory and message passing architectures and is built as a cluster of SMP workstations using 4-way SMP workstations from Sun Microsystems. The work discusses about Parallel Finite element Analysis Programs developed with Direct Frontal Solver and Iterative PCG (Pre-conditioned Conjugate Gradient Method) Solver for 2d and 3d analysis. The Comparative Performance of these Parallel Computation of FEA (Direct and Iterative Solver based) is bench marked on PARAM-10000. Studies have been conducted on medium sized two and three dimensional linear finite element analysis problems for elastic structural analysis. To increase problem size(dof), automated meshes were generated using 8-noded serendipity elements for 2d analysis and 20-noded brick elements for 3d analysis. Each problem has been analysed by Parallel FEA Computation using 1 to 4 Processors on PARAM-10000. Detailed Speed-up and Efficiency studies have been reported and conclusions are drawn about Performance on PARAM-10000 machine.