Advertisement

Study of Load Balancing Strategies for Finite Element Computations on Heterogeneous Clusters

  • Kalyani Munasinghe
  • Richard Wait
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3732)

Abstract

We study strategies for redistributing the load in an adaptive finite element computation performed on a cluster of workstations. The cluster is assumed to be a heterogeneous, multi-user computing environment. The performance of a particular processor depends on both static factors, such as the processor hardware and dynamic factors, such as the system load and the work of other users.

On a network, it is assumed that all processors are connected, but the topology of the finite element sub-domains can be interpreted as a processor topology and hence for each processor, it is possible to define set of neighbours. In finite element analysis, the quantity of computation on a processor is proportional to the size of the sub-domain plus some contribution from the neighbours. We consider schemes that modify the sub-domains by, in general, moving data to adjacent processors. The numerical experiments show the efficiency of the approach.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
    Bevilacqua, A.: A dynamic load balancing method on a heterogeneous cluster of workstations. Informatica 23(1), 49–56 (1999)Google Scholar
  5. 5.
    Cybenko, G.: Dynamic load balancing for distributed memory multiprocessors. Parallel and Distributed Computing 7, 279–301 (1989)CrossRefGoogle Scholar
  6. 6.
    Lan, Z., Taylor, V.E.: Dynamic load balancing of SAMR applications on distributed systems. Scientific Programming 10(21), 319–328 (2002)Google Scholar
  7. 7.
    Lee, C.K., Hamdi, M.: Parallel image processing application on a network of distributed workstations. Parallel Computing 26, 137–160 (1995)CrossRefGoogle Scholar
  8. 8.
    Lin, J., Saletore, V.A.: Self scheduling on distributed memory machines. Supercomputing, 814–823 (1993)Google Scholar
  9. 9.
    Oliker, L., Biswas, R.: Plum: Parallel load balancing for adaptive structured meshes. Parallel and Distributed Computing 52(2), 150–177 (1998)zbMATHCrossRefGoogle Scholar
  10. 10.
    Schloegel, K., Karypis, G., Kumar, V.: Multilevel diffusion schemes for repartitioning of adaptive meshes. Journal of Parallel and Distributed Computing 47(2), 109–124 (1997)CrossRefGoogle Scholar
  11. 11.
    Weissman, J.B.: Predicting the Cost and Benefit of Adapting Data Parallel Applications in Clusters. Parallel and Distributed Computing 62, 1248–1271 (2002)zbMATHCrossRefGoogle Scholar
  12. 12.
    Horton, G.: A multilevel diffusion method for dynamic load balancing. Parallel Computing 9, 209–218 (1993)CrossRefGoogle Scholar
  13. 13.
    Schloegel, K., Karypis, G., Kumar, V.: Multilevel Diffusion Schemes for Repartitioning of Adaptive Meshes. Journal of Parallel and Distributed Computing 47(2), 109–124 (1997)CrossRefGoogle Scholar
  14. 14.
    Walshaw, C., Cross, M., Everett, M.G.: ParallelDynamicGraph Partitioning forAdaptive Unstructured Meshes. Journal of Parallel and Distributed Computing 47(2), 102–108 (1997)CrossRefGoogle Scholar
  15. 15.
    Hu, Y.F., Blake, R.J.: An improved diffusion algorithmfor dynamic load balancing. Parallel Computing 25(4), 417–444 (1999)zbMATHCrossRefMathSciNetGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Kalyani Munasinghe
    • 1
  • Richard Wait
    • 2
  1. 1.Dept. of Computer ScienceUniversity of RuhunaSri Lanka
  2. 2.Dept. of Scientific ComputingUppsala UniversitySweden

Personalised recommendations