A GPU-accelerated Navier-Stokes Solver for Steady Turbomachinery Simulations

Any tiny improvement of modern turbomachinery components require nowadays a large amount of design evaluations. Every evaluation runs time consuming simulations. Reducing the computational cost of the simulations allows to run more evaluations, thus reaching a higher design improvement. In this work, an Nvidia Graphics Processing Unit (GPU) of Kepler generation is used to accelerate the FiniteVolume CFD flow evaluation as part of an inhouse multidisciplinary design optimization (MDO) algorithm. The validated double precision GPU solver is up to 30x times faster for inviscid applications and up to 10 times faster for turbulent viscous cases than a the serial code running on an Intel(R) Xeon(R) CPU E3-1240 v3. The gain on the CFD simulation time make it possible to run the MDO optimizer with explicit solver on a workstation under affordable time budget.