A Practical Approach to Efficient Use of Heterogeneous PC Network for Parallel Mathematical Computation
This paper presents an experience in use of a parallel mathematical library, ScaLAPACK, on a network composed by heterogeneous workstations. The good performance results have been obtained by means of a distributed programming environment which is able to dynamically evaluate available computing power at each workstation and to distribute accordingly the set of parallel processes.
KeywordsTarget Architecture Static Tiling Good Performance Result Basic Linear Algebra Host Node
Unable to display preview. Download preview PDF.
- 4.M. Cermele, M. Colajanni, G. Necci, “Dynamic load balancing of distributed SPMD computations with explicit message-passing”, Proc. 6th Heterogeneous Computing Workshop, pp. 2–16, 1997.Google Scholar
- 5.A. Clematis G. Dodero, V. Gianuzzi, “A resource management tool for heterogeneous networks”, Proc. 7th Euromicro Workshop on Parallel and Distributed Processing, IEEE Comp. Soc. Press, pp. 367–373, 1999.Google Scholar
- 6.G. Dodero, V. Gianuzzi, M. Moscati, M. Corvi, “A Scalable Parallel Algorithm for Matching Pursuit Signal Decomposition”, Proc. HPCN’98 Conference, Lect. Notes in Comp. Sciences, pp. 458–466, 1998.Google Scholar