A Mesh Partitioning Strategy for Optimizing the Performance of Adaptive Explicit Time-Stepping Methods

Explicit time-stepping methods are commonly used to solve transient, timedependent problems. For problems involving small-scale features, adaptive unstructured mesh refinement can be effective. Unfortunately, using such dynamic strategies on parallel computers can result in difficult load balancing issues because of the relatively high cost of the mesh partitioning to the time-step computation. We present a partitioning strategy that addresses this problem and demonstrate the effectiveness of the proposed method in modeling of a two-dimensional Raleigh-Taylor instability.