Abstract: Hierarchical Partitioning and Dynamic Load Balancing for Scientific Computation

Hierarchical Partitioning and Dynamic Load Balancing for Scientific Computation

James D. Teresco, Jamal Faik, and Joseph E. Flaherty
Proc. Workshop on State-Of-The-Art in Scientific Computing: 7th International Conference, PARA 2004, Lyngby, Denmark, June 20-23, 2004. Jack Dongarra, Kaj Madsen, Jerzy Wasniewski, editors. Lecture Notes in Computer Science 3732, pp. 911-920, © Springer-Verlag. 2006. Previously published as Williams College Department of Computer Science Technical Report CS-04-04, and Sandia Report SAND2004-1559A, Sandia National Laboratories, 2004.

Cluster and grid computing has made hierarchical and heterogeneous computing systems increasingly common as target environments for large-scale scientific computation. A cluster may consist of a network of multiprocessors. A grid computation may involve communication across slow interfaces. Modern supercomputers are often large clusters with hierarchical network structures. For maximum efficiency, software must adapt to the computing environment. We focus on partitioning and dynamic load balancing, in particular on hierarchical procedures implemented within the Zoltan Toolkit, guided by DRUM, the Dynamic Resource Utilization Model. Here, different balancing procedures are used in different parts of the domain. Preliminary results show benefits to using hierarchical partitionings on hierarchical systems.

Citation (BIBTEX)  Extended Abstract (PDF; 87KB)  Paper (Tech Report version) (PDF; 224KB)  Published version (LNCS web site)