Study of Incompressible Flows in Rectangular Channels Using High Order Schemes and Parallel Computing

Abstract

Results from the parallelization of a three-dimensional incompressible Navier-Stokes programme are presented. The programme is based on the artificial compressibility method for coupling the continuity with the momentum equations. The convection terms are discretized by a third order upwind characteristic based method. Second order central differences are used for the viscous terms. The time integration is obtained by an explicit Runge-Kutta scheme. The Navier-Stokes method has been parallelized using shared memory as well as message-passing models. Computations have been performed for a straight three-dimensional channel of rectangular cross-section. The efficiency of the parallel models is presented for different grid sizes and number of processors.KeywordsShared MemoryIncompressible FlowTotal EfficiencyHigh Order SchemeParallelization StrategyThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.