Massively parallel implementation of an explicit CFD algorithm on unstructured grids

In this paper we present the implementation of a new CFD algorithm on massively parallel architectures. First, the numerical method itself is reviewed, with emphasis on its salient features. Next, we show how the algorithm can be ported to parallel architectures using the synchronous, hosted model. We conclude by showing several 2-D and 3-D computational results, some of which are of real life complexity.