Large calculation of the flow over a hypersonic vehicle using a GPU

Graphics processing units are capable of impressive computing performance up to 518Gflops peak performance. Various groups have been using these processors for general purpose computing; most efforts have focussed on demonstrating relatively basic calculations, e.g. numerical linear algebra, or physical simulations for visualization purposes with limited accuracy. This paper describes the simulation of a hypersonic vehicle configuration with detailed geometry and accurate boundary conditions using the compressible Euler equations. To the authors' knowledge, this is the most sophisticated calculation of this kind in terms of complexity of the geometry, the physical model, the numerical methods employed, and the accuracy of the solution. The Navier-Stokes Stanford University Solver (NSSUS) was used for this purpose. NSSUS is a multi-block structured code with a provably stable and accurate numerical discretization which uses a vertex-based finite-difference method. A multi-grid scheme is used to accelerate the solution of the system. Based on a comparison of the Intel Core 2 Duo and NVIDIA 8800GTX, speed-ups of over 40x were demonstrated for simple test geometries and 20x for complex geometries.

[1]  Anselmo Lastra,et al.  Simulation of cloud dynamics on graphics hardware , 2003, HWWS '03.

[2]  J. Gong,et al.  A Hybrid Method for the Unsteady Compressible Navier-Stokes Equations , 2007 .

[3]  Arie E. Kaufman,et al.  Implementing lattice Boltzmann computation on graphics hardware , 2003, The Visual Computer.

[4]  Martin Rumpf,et al.  Nonlinear Diffusion in Graphics Hardware , 2001, VisSym.

[5]  Juan J. Alonso,et al.  High-Speed MHD Flow Control Using Adjoint-Based Sensitivities , 2006 .

[6]  Jens H. Krüger,et al.  A Survey of General‐Purpose Computation on Graphics Hardware , 2007, Eurographics.

[7]  Magnus Svärd,et al.  Accuracy requirements for transient aerodynamics , 2003 .

[8]  Carlos Eduardo Scheidegger,et al.  Practical CFD Simulations on Programmable Graphics Hardware using SMAC † , 2005, Comput. Graph. Forum.

[9]  Arie E. Kaufman,et al.  GPU Cluster for High Performance Computing , 2004, Proceedings of the ACM/IEEE SC2004 Conference.

[10]  Liu You,et al.  Real-Time 3D Fluid Simulation on GPU with Complex Obstacles , 2006 .

[11]  J. Krüger,et al.  Linear algebra operators for GPU implementation of numerical algorithms , 2003, ACM Trans. Graph..

[12]  Graham Pullan,et al.  Acceleration of a 3D Euler solver using commodity graphics hardware , 2008 .

[13]  Naga K. Govindaraju,et al.  GPGPU: general-purpose computation on graphics hardware , 2006, SC.

[14]  D. Gottlieb,et al.  Time-stable boundary conditions for finite-difference schemes solving hyperbolic systems: methodology and application to high-order compact schemes , 1994 .

[15]  Jos Stam,et al.  Stable fluids , 1999, SIGGRAPH.

[16]  Magnus Svärd,et al.  Steady-State Computations Using Summation-by-Parts Operators , 2005, J. Sci. Comput..

[17]  Greg Humphreys,et al.  A multigrid solver for boundary value problems using programmable graphics hardware , 2003, HWWS '03.

[18]  Pat Hanrahan,et al.  Brook for GPUs: stream computing on graphics hardware , 2004, SIGGRAPH 2004.

[19]  Arie E. Kaufman,et al.  GPU-Based flow simulation with complex boundaries , 2010 .

[20]  Ian Buck High level languages for GPUs , 2005, SIGGRAPH Courses.

[21]  A. A. Amsden,et al.  The SMAC method: A numerical technique for calculating incompressible fluid flow , 1970 .