PERFORMANCE ANALYSIS OF VISCOUS-FLOW COMPUTATIONS ON VARIOUS PARALLELARCHITECTURES

A study on the performance of parallel viscous flow computations on va rious parallel architectures is presented. An implicit computational f luid dynamics code was utilized to solve the axisymmetric compressible Navier-Stokes equations, and parallelization of the code was obtained by the grid partitioning technique. Several hardware and software iss ues, such as, different communication possibilities and their effect o n the parallel performance, the numerical and parallel efficiency for single grid and mesh-sequencing solutions, double and single precision calculations, and the effect of the number of processors on the compu ting rime and the parallel performance, were investigated. Calculation s for laminar and turbulent benchmark cases were performed on four par allel platforms and the results were compared with corresponding compu tations done on the Cray Y-MP utilizing only one processor. A theoreti cal model for the prediction of the parallel efficiency was also deriv ed and it was verified for single grid and mesh-sequencing solutions b y comparing the measured efficiencies with those of the predicted ones .