Distributed computing in a heterogeneous computing environment
Distributed computing is a means to overcome the limitations of single computing systems. In this paper we describe how clusters of heterogeneous supercomputers can be used to run a single application or a set of applications. We concentrate on the communication problem in such a configuration and present a software library called PACX-MPI that was developed to allow a single system image from the point of view of an MPI programmer. We describe the concepts that have been implemented for heterogeneous clusters of this type and give a description of real applications using this library.
Unable to display preview. Download preview PDF.
- 1.Th. Eickermann, J. Heinrichs, M. Resch, R. Stoy, R. Völpel,’ Metacomputing in Gigabit Environments: Networks, Tools and Appplications’ to appear in Parallel Computing (1998).Google Scholar
- 2.Toshiya Kimura, Hiroshi Takemiya,’ Local Area Metacomputing for Multidisciplinary Problems: A Case Study for Fluid/Structure Coupled Simulation', 12th ACM International Conference on Supercomputing, Melbourne, July 13–17 (1998).Google Scholar
- 3.Thomas Bönisch and Roland Rühle,’ Adapting a CFD code for metacomputing', 10th International Conference on Parallel CFD, Hsinchu/Taiwan, May 11–14, (1998).Google Scholar
- 4.W. Gropp, E. Lusk, N. Doss, A. Skjellum,’ A high-performance, portable implementation of the MPI message-passing interface standard', Parallel Computing, 22 (1996).Google Scholar
- 5.Graham E. Fagg, Jack J. Dongarra and Al Geist,’ Heterogeneous MPI Application Interoperation and Process management under PVMPI', in: Marian Bubak, Jack Dongarra, Jerzy Wasniewski (Eds.),’ Recent Advances in Parallel Virtual Machine and Message Passing Interface', 91–98, Springer (1997).Google Scholar
- 7.Thomas Beisel, Edgar Gabriel, Michael Resch,’ An Extension to MPI for Distributed Computing on MPPs' in: Marian Bubak, Jack Dongarra, Jerzy Wasniewski (Eds.)’ Recent Advances in Parallel Virtual Machine and Message Passing Interface', Lecture Notes in Computer Science, 75–83, Springer (1997).Google Scholar
- 8.Matthias Müller and Hans J. Herrmann,’ DSMC — a stochastic algorithm for granular matter', in: Hans J. Herrmann and J.-P. Hovi and Stefan Luding (Eds.)’ Physics of dry granular media', Kluwer Academic Publisher (1998).Google Scholar