Computing Boundary Element Method’s Matrices on GPU
Conference paper
Abstract
Matrices resulting from standard boundary element methods are dense and computationally expensive. To speed up the computational time, the matrix computation is done on a GPU. The parallel processing capability of the Graphics Processing Unit (GPU) allows us to divide complex computing tasks into several thousands of smaller tasks that can be run concurrently. We achieved an acceleration of 31 − 36 in comparison to a computation performed on the CPU, serially.
Keywords
Graphic Processing Unit Boundary Element Method Shared Memory Collocation Point Global Memory
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Preview
Unable to display preview. Download preview PDF.
References
- 1.Schanz, M.: Wave Propagation in Viscoelastic and Poroelastic Continua: A Boundary Element Approach. Lecture Notes in Applied Mechanics, vol. 2 (2001)Google Scholar
- 2.Guiggiani, M., Gigante, A.: A general algorithm for multidimensional Cauchy principal value integrals in the boundary element method. J. of Appl. Mech. 57, 906–915 (1990)MathSciNetMATHCrossRefGoogle Scholar
- 3.Schanz, M.: On a Reformulated Convolution Quadrature Based Boundary Element Method. Computer Modeling in Engineering and Sciences 58, 109–128 (2010)MathSciNetMATHGoogle Scholar
- 4.Dunavant, D.A.: High degree efficient symmetrical Gaussian quadrature rules for the triangle. IJNME 21, 1129–1148 (1985)MathSciNetMATHCrossRefGoogle Scholar
- 5.NVIDIA Corporation, NVIDIA CUDA C Programming Guide Version 4.0 (June 2011), http://developer.download.nvidia.com/compute/DevZone/docs/html/C/doc/CUDA_C_Programming_Guide.pdf
Copyright information
© Springer-Verlag Berlin Heidelberg 2012