Finite Element Matrix Generation on a GPU
Wireless Communication Engineering (WiComm) Center of Excellence, Department of Microwave and Antenna Engineering, Faculty of Electronics, Telecommunications and Informatics, CUDA Research Center for Computational Electromagnetics, Gdansk University of Technology, Gdansk 80-233, Poland
Progress In Electromagnetics Research, Vol. 128, 249-265, 2012
@article{dziekonski2012finite,
title={Finite Element Matrix Generation on a GPU},
author={Dziekonski, A. and Sypek, P. and Lamecki, A. and Mrozowski, M.},
year={2012}
}
This paper presents an efficient technique for fast generation of sparse systems of linear equations arising in computational electromagnetics in a finite element method using higher order elements. The proposed approach employs a graphics processing unit (GPU) for both numerical integration and matrix assembly. The performance results obtained on a test platform consisting of a Fermi GPU (1x Tesla C2075) and a CPU (2x twelve-core Opterons), indicate that the GPU implementation of the matrix generation allows one to achieve speedups by a factor of 81 and 19 over the optimized single-and multi-threaded CPU-only implementations, respectively.
June 4, 2012 by hgpu
