Implementation and analysis of Jacobi iteration based on hybrid programming
Anlei Shi, Weichang Shen, Yongjin Li, Lidong He, Zhao Dong · 2010
With the development of high-speed networks and the multi-core processor technology, the cluster of workstation based on high-speed networks and multi-core processors is becoming the main platform for parallel computing. Jacobi iterative method for solving linear equations is a common method, there are widely range of applications in many areas of science and engineering. This paper parallelizes Jacobi iterative method in process-level using MPI at first, and identifies the most time-consuming part of the program, then parallelizes in thread-level using OpenMP based on shared memory, so that the program can take full advantage of multi-core workstations to reduce the computation time.