The preconditioned conjugate gradient method on distributed memory systems
We present an implementation method for the preconditioned conjugate gradient algorithm with geometric domain decomposition. The results of some experiments on both a 4 processor- and a 512 processor system are presented and expectations for the implementation on more processors are discussed.
Unable to display preview. Download preview PDF.
- 1.E. F. D'Azevedo and C. H. Romine, Reducing communication costs in the conjugate gradient algorithm on distributed memory multiprocessors, Oak Ridge National Laboratory, ORNL/TM-12192, sept. 1992.Google Scholar
- 2.A. T. Chronopoulos, C. W. Gear, S-step iterative methods for symmetric linear systems, Journal of Computational and Applied Mathematics 25 (1989), p. 153–168.Google Scholar
- 3.J. W. Demmel, M. T. Heath and H. A. van der Vorst, Parallel numerical linear algebra, Acta Numerica, Vol. 2, 1993, p. 111–199.Google Scholar
- 4.J. J. Dongarra, I. S. Duff, D. C. Sorenson and H. A. van der Vorst, Solving linear systems on vector and shared memory computers, SIAM, Philadelphia, 1991.Google Scholar
- 5.Y. Saad, Krylov subspace methods on supercomputers, SIAM Journal on scientific and statistical computing 10 vol 2 (1989), p. 1200–1232.Google Scholar
- 6.E. de Sturler, A parallel restructured version of GMRES(m), technical report 91-85, University of Delft, Delft, 1991.Google Scholar