A PARALLEL PROGRAMMING ENVIRONMENT FOR ADAPTIVE P-VERSION FINITE-ELEMENT ANALYSIS

This paper presents a parallel programming environment for solving lar ge-scale problems using adaptive p-version finite element approach. A scalable parallel feedback algorithm based on domain decomposition tec hnique has been developed to incorporate the parallelism in adaptive f inite element analysis of large-scale structures. This algorithm is im plementable on MIMD type parallel processors, and uses p-extension of the finite element method. Modeling with p-version finite elements nee ds special considerations of load-balancing, and this has been incorpo rated in present domain decomposition technique using semi-empirical a pproach. The Connection Machine's CM-5 system has been used to impleme nt the present system. Most of the previous attempts to parallelize ad aptive finite element analysis were confirmed to parallel algorithms f or direct or iterative equation solvers, and did not produce satisfact ory performance because of small granularity of parallel tasks. Also t hese efforts were based on h-version of the finite element method, and hence could not exploit some of the special advantages available in p -vision of finite element approach. As the feedback algorithm is based on an iterative scheme, utilizing the domain decomposition technique, it produced good performance, particularly for large-scale structures . It has been tested for a number of regular and irregular two-dimensi onal problems showing good convergence characteristics. (C) 1998 Elsev ier Science Limited. All rights reserved.