Comparison of message-passing and shared memory implementations of the GMRES method on MIMD computers

摘要

In this paper we compare different parallel implementations of the same algorithm for solving nonlinear simulation prob- lems on unstructured meshes. In the first implementation, making use of the message- Passing programming model and the PVM system, the do- Main decomposition of unstructured mesh is implemented, While the second implementation takes advantage of the in- Herent parallelism of the algorithm by adopting the shared- Memory programming model. Both implementations are ap- Plied to the preconditioned GMRES method that solves itera- Tively the system of linear equations.