| Author | Joseph E. Flaherty and James D. Teresco |
|---|---|
| Title | Architecture-Dependent Load Balancing Using the Rensselaer Partition Model |
| Year | 1999 |
| Abstract | Reusable software tools for the solution of partial differential equations using parallel adaptive finite element methods have been developed. We describe the design and implementation of object-oriented parallel mesh structures within an adaptive framework called Trellis. The Rensselaer Partition Model (RPM) is central to the framework. This hierarchical model is used to distribute finite element meshes and associated data on a parallel computer. It represents heterogeneous processor and network speeds, and may be used to represent processes in any parallel computing environment, including an SMP, a distributed-memory computer, a network of workstations, or some combination of these. The information about different processor speeds, memory sizes, and the corresponding interconnection network can be useful in a dynamic load balancing algorithm which seeks to achieve a good balance with minimal interprocessor communication penalties when a slow interconnection network is involved. An example of a partitioning scheme which takes advantage of this information is given. |