An Application-Centric Characterization of Distribution Techniques for Dynamic Adaptive Grid Hierarchies

Dynamically adaptive methods for the solution of partial differential equations that employ locally optimal approximations can yield highly advantageous ratios for cost/accuracy. Distributed implementations of these methods offer the potential for accurate solution of physically realistic models of important physical systems. These implementations however, lead to interesting challenges in dynamic data-distribution and load balancing. This paper presents ongoing work on characterizing the performance of dynamic partitioning and loadbalancing techniques for distributed adaptive grid hierarchies that underlie adaptive mesh-refinement algorithms (AMR). The overall goal of this characterization is to enable the selection of the most appropriate mechanism based on application and system parameters.