International Conference on Parallel Computing in Electrical Engineering (PARELEC'02) (2002)
Sept. 22, 2002 to Sept. 25, 2002
Robert Piotr Bycul , Technical University of Bialystok
Andrzej Jordan , Technical University of Bialystok
Marcin Cichomski , Polish-Japanese Institute of Information Technology
<p>In the article the authors describe an idea of parallel implementation of a conjugate gradient method in a heterogeneous PC cluster and a supercomputer Hitachi SR-2201. The new version of algorithm implementation differs from the one applied earlier , because it uses a special method for storing sparse coefficient matrices: only non-zero elements are stored and taken into account during computations, so that the sparsity of the coefficient matrix is taken full advantage of.</p> <p>The article includes a comparison of the two versions. A speedup of the parallel algorithm has been examined for three different cases of coefficient matrices resulting in solving different physical problems. The authors have also investigated a preconditioning method, which uses the inversed diagonal of the coefficient matrix, as a preconditioning matrix.</p>
A. Jordan, M. Cichomski and R. P. Bycul, "A New Version of Conjugate Gradient Method Parallel Implementation," International Conference on Parallel Computing in Electrical Engineering (PARELEC'02)(PARELEC), Warsaw, Poland, 2002, pp. 318.