论文信息 - On the acceleration of the numerical solution of partial differential equations using radial basis functions and graphics processing units

On the acceleration of the numerical solution of partial differential equations using radial basis functions and graphics processing units

General purpose computation on graphics processing units is becoming very important as a means for speeding up numerical computations. In this research, nVidia's Compute Unified Device Architecture (CUDA) language is used along with a graphics processing unit to accelerate the generation of the system matrix for the numerical solution of a partial differential equation employing radial basis functions. Details of the CUDA implementation as well as speedup curves using several different graphics processing units will be discussed. Copyright © 2012 John Wiley & Sons, Ltd.

Richard K. Gordon | W. Elliott Hutchcraft | Maxwell Woolsey

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

[2] E. Kansa,et al. Exponential convergence and H‐c multiquadric collocation method for partial differential equations , 2003 .

[3] Ioannis T. Rekanos. On-line inverse scattering of conducting cylinders using radial basis-function neural networks , 2001 .

[4] Liu Guosui,et al. Radar target classification based on radial basis function and modified radial basis function networks , 1996, Proceedings of International Radar Conference.

[5] David Tarditi,et al. Accelerator: using data parallelism to program GPUs for general-purpose uses , 2006, ASPLOS XII.

[6] Satish S. Udpa,et al. Solution of inverse problems in electromagnetics using Hopfield neural networks , 1995 .

[7] Salvatore Caorsi,et al. A threshold electromagnetic classification approach for cylinders embedded in a lossy medium by using a neural network technique , 2000 .

[9] Alexander I. Fedoseyev,et al. Continuation for nonlinear Elliptic Partial differential equations discretized by the multiquadric Method , 2000, Int. J. Bifurc. Chaos.

[10] Krzysztof A. Michalski,et al. A neural-network approach to the electromagnetic imaging of elliptic conducting cylinders , 2001 .