An Efficient MPI Virtual Topology Based Parallel, Iterative MoM-PO Hybrid Method on PC Clusters

In this paper, a novel iterative MoM-PO hybrid method is implemented by iterating the voltag e matrix of the method of moments. To make the MoM-PO code more applicable to real world problems, an efficient load-balancing MPI based parallel MoM-PO hybrid scheme is further described in detail. The introducing of the MPI virtual topology for MoM-PO method makes the parallelization more convenient and efficient. Several numerical results involving complex geometries such as an aircraft have been presented to demonstrate the accuracy and efficiency of the parallel MoM-PO method.