A Simple Method for Static Load Balancing of Parallel FDTD Codes

摘要

A static method for balancing computational loads in parallel implementations of the finite-difference timedomain method is presented. The procedure is fairlystraightforward and computationally inexpensive, thus providing an attractive alternative to optimization techniques. The method is described for partitioning in a single mesh dimension, but it is shown that it can be adapted also for 2D and 3D partitioning in approximate way, with good results. It is applicable to both homogeneous and heterogeneous parallel architectures, and can also be used for balancing memory on distributed memory architectures.