Parallel partitioning of unstructured meshes

Publisher Summary

The use of unstructured mesh codes on parallel machines can be one of the most efficient ways to solve large computational-fluid-dynamics (CFD) and computational-mechanics (CM) problems. An important consideration, however, is the problem of distributing the mesh across the memory of the machine at run time so that the computational load is evenly balanced and the amount of interprocessor communication is minimized. The algorithms can also be used for dynamic load balancing and a clustering technique can additionally be employed to speed up the whole process. The idea behind the subdomain heuristic is to minimize the surface energy of the subdomain. For coarse granularity partitions, it is inefficient to apply the optimization techniques to every graph vertex as most will be internal to the sub domains. A simple technique to speed up the optimization process, therefore, is to group vertices together to form clusters, use the clusters to define a new graph, recursively iterate this procedure until the graph size falls below some threshold, and then apply the partitioning algorithm to these reduced size graphs.