The Community for Technology Leaders
RSS Icon
Issue No.02 - March/April (2009 vol.11)
pp: 52-59
George Stantchev , University of Maryland, College Park
Derek Juba , University of Maryland, College Park
William Dorland , University of Maryland, College Park
Amitabh Varshney , University of Maryland, College Park
Direct numerical simulation (DNS) of turbulence is computationally intensive and typically relies on some form of parallel processing. Spectral kernels used for spatial discretization are a common computational bottleneck on distributed memory architectures. One way to increase DNS algorithms' efficiency is to parallelize spectral kernels using tightly coupled single-program, multiple-data (SPMD) multiprocessor units with minimal interprocessor communication latency. The authors present techniques to map DNS computations to modern graphics processing units (GPUs), which are characterized by very high memory bandwidth and hundreds of SPMD processors. The article compares the performance between the authors' parallel algorithm running on a GPU versus the associated CPU implementation of a solver for one of the fundamental nonlinear models of turbulence theory. They also demonstrate a prototype of a scalable computational steering framework based on turbulence simulation and visualization coupling on the GPU.
GPU programming, scientific computing, direct numerical simulation, plasma turbulence, graphics processing unit
George Stantchev, Derek Juba, William Dorland, Amitabh Varshney, "Using Graphics Processors for High-Performance Computation and Visualization of Plasma Turbulence", Computing in Science & Engineering, vol.11, no. 2, pp. 52-59, March/April 2009, doi:10.1109/MCSE.2009.42
1. J.D. Owens et al., "A Survey of General Purpose Computation on Graphics Hardware," Computer Graphics Forum, vol. 26, no. 1, pp. 80–113.
2. Nvidia, "CUDA Developers Zone," 2008;
3. M. Yokokawa et al., "16.4-Tflops Direct Numerical Simulation of Turbulence by a Fourier Spectral Method on the Earth Simulator," Proc. 2002 ACM/IEEE Conf. Supercomputing (SC 02), IEEE CS Press, 2002, pp. 1–17.
4. G.G. Howes et al., "Astrophysical Gyrokinetics: Basic Equations and Linear Theory," Astrophysical J., vol. 651, no. 590, 2006, pp. 590–614.
5. W. Dorland et al., "Electron Temperature Gradient Turbulence," Physics Rev. Letters, vol. 85, 2000, pp. 5579–5582.
6. A. Hasegawa and K. Mima,"Stationary Spectrum of Strong Turbulence in Magnetized Nonuniform Plasma," Physics Rev. Letters, vol. 39, no. 205, 1977.
7. N.K. Govindaraju et al., "High-Performance Discrete Fourier Transforms on Graphics Processors," Proc. 2008 ACM/IEEE Conf. Supercomputing (SC 08), IEEE Press, 2008, pp. 1–12.
8. P.J. Morrison and R.D. Hazeltine,"Hamiltonian Formulation of Reduced Magnetohydrodynamics," Physics of Fluids, vol. 27, no. 4, 1984, pp.886–897.
9. A.A. Schekochihin et al., "Kinetic and Fluid Turbulent Cascades in Magnetized Weakly Collisional Astrophysical Plasmas," ArXiv e-prints, 2007;
10. S.E. Parker and W.W. Lee., "A Fully Nonlinear Characteristic Method for Gyrokinetic Simulation," Physics of Fluids B, vol. 5, no. 77, 1993, pp. 77–86.
11. X. Wei et al., "The Lattice-Boltzmann Method for Simulating Gaseous Phenomena," IEEE Trans. Visualization and Computer Graphics, vol. 10, no. 2, 2004, pp. 164–176.
14 ms
(Ver 2.0)

Marketing Automation Platform Marketing Automation Tool