Computing thr minimum eigenvalue of a symmetric positive definite Toeplitz matrix
SIAM Journal on Scientific and Statistical Computing
Parallel Programming: Techniques and Applications Using Networked Workstations and Parallel Computers (2nd Edition)
Hi-index | 0.00 |
In this paper MPI is used on PC Cluster to compute all the eigenvalues of Hermitian Toeplitz Matrices The parallel algorithms presented were implemented in C++ with MPI functions inserted and run on a cluster of Lenovo ThinkCentre machines running RedHat Linux The two methods, MAHT-P one embarrassingly parallel and the other MPEAHT using master/ slave scheme are compared for performance and results presented It is seen that computation time is reduced and speedup factor increases with the number of computers used for the two parallel schemes presented Load balancing becomes an issue as number of computers in a cluster are increased A solution is provided to overcome such a case.