Dynamic Load Balancing for Parallel Numerical Simulations Based on Repartitioning with Disturbed Diffusion

Load balancing is an important requirement for the efficient execution of parallel numerical simulations. In particular when the simulation domain changes over time, the mapping of computational tasks to processors needs to be modified accordingly. State-of-the-art libraries for this problem are based on graph repartitioning. They have a number of drawbacks, including the optimized metric and the difficulty of parallelizing the popular repartitioning heuristic Kernighan-Lin (KL). In this paper we further explore the very promising diffusion-based graph partitioning algorithm DIBAP by adapting DIBAP to the related problem of load balancing and improving its implementation. The presented experiments with graph sequences that imitate adaptive numerical simulations are the first using DIBAP for load balancing. They demonstrate the applicability and high quality of DIBAP for load balancing by repartitioning. Compared to the faster state-of-the-art repartitioners PARMETIS and parallel JOSTLE, DIBAP´s solutions have partitions with significantly fewer external edges and boundary nodes and the resulting average migration volume in the important maximum norm is also the best in most cases.