Mesh partitioning for distributed systems
Mesh partitioning for homogeneous systems has been studied extensively; however, mesh partitioning for distributed systems is a relatively new area of research. To ensure efficient execution on a distributed system, the heterogeneities in the processor and network performance must be taken into cons...
Uložené v:
| Hlavný autor: | |
|---|---|
| Médium: | Dissertation |
| Jazyk: | English |
| Vydavateľské údaje: |
ProQuest Dissertations & Theses
01.01.1999
|
| Predmet: | |
| ISBN: | 0599325836, 9780599325838 |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Shrnutí: | Mesh partitioning for homogeneous systems has been studied extensively; however, mesh partitioning for distributed systems is a relatively new area of research. To ensure efficient execution on a distributed system, the heterogeneities in the processor and network performance must be taken into consideration in the partitioning process; equal size subdomains and small cut set size, which results from conventional mesh partitioning, are no longer the primary goals. In this thesis, we address various issues related to mesh partitioning for distributed systems. These issues include the metric used to compare different partitions, efficiency of the application executing on a distributed system, and the advantage of exploiting heterogeneity in network performance. We present a tool called PART, for automatic mesh partitioning for distributed systems. The novel feature of PART is that it considers heterogeneities in the application and the distributed system. Simulated annealing is used in PART to perform the backtracking search for desired partitions. While it is well-known that simulated annealing is computationally-intensive, we describe the parallel version of simulated annealing that is used with PART. The results of the parallelization exhibit nearly perfect speedup in most cases and superlinear speedup in some cases. Experimental results are presented for partitioning regular and irregular finite element meshes for an explicit finite element application executing on a geographically distributed system consisting of two IBM SPs. The results from the regular problems indicate a 33–46% increase in efficiency when processor performance is considered as compared to the conventional even partitioning. The results indicate 5–15% increase in efficiency when network performance is considered as compared to considering only processor performance; this is significant given that the optimal is 15% for this application for up to 8 processors. The results from the irregular problems indicate up to 35% increase in efficiency as the number of processors increase to 40 when network performance is considered as compared to only consider processor performance. Preliminary experimental results are also presented for an implicit finite element application. The results indicate up to 15% improvement in efficiency when network performance is taken into consideration as compared to only consider processor performance. |
|---|---|
| Bibliografia: | SourceType-Dissertations & Theses-1 ObjectType-Dissertation/Thesis-1 content type line 12 |
| ISBN: | 0599325836 9780599325838 |