Parallel implementations of the boundary element method

The boundary element method has its origins in the boundary integral equation method [1] and has, in the past two decades, become a well-established technique for the solution of problems in engineering and applied science. A significant amount of work has been focused on the development of computer programs, the vast majority of which have been written to run on sequential computers. However, the boundary element method exhibits inherent parallelisms which may be mapped onto a variety of parallel architectures. Over the past few years, boundary element researchers have started to realize the possibilities that parallel computing environments offer. Applications from elasticity, electromagnetics, fluid dynamics, etc., have had implementations on both fine-grained and coarse-grained architectures.