Implicit residual smoothing in a parallel 2D explicit Euler solver

This paper deals with the parallel implementation on distributed memory arc hitectures of the implicit residual smoothing procedure in the context of a explicit method for two dimensional inviscid flows. The governing equation s are discretized by a cell centered finite volume method and the time inte gration is performed by a explicit Runge Kutta method. Artificial dissipati on and implicit residual smoothing are used in order to stabilize and speed up the method. The parallelism is introduced by grid partitioning. The par allel implementation of the residual smoothing, a inherently implicit proce dure, is crucial for the efficiency of the method. Here, two different para llel residual smoothing strategies are discussed and some experimental resu lts are given to illustrate parallel performances of the proposed strategie s.