论文信息 - Parallelizing multilevel fast multipole algorithm for large-scale electromagnetic problem on GPU clusters

Parallelizing multilevel fast multipole algorithm for large-scale electromagnetic problem on GPU clusters

In this paper, we investigate the parallelization of the Multi-level Fast Multipole Algorithm (MLFMA) for solving large scale electromagnetic scattering problems. We evaluate the performance on our 13-node Graphics Processing Unit (GPU) cluster supported by Nvidia Tesla M2090 GPUs. The implementation details and the performance achievements in terms of accuracy are discussed in this paper.

Ozlem Kilic | Tuan Phan | Nghia Tran

[1] Vinh Dang,et al. Parallelizing Fast Multipole Method for Large-Scale Electromagnetic Problems Using GPU Clusters , 2013, IEEE Antennas and Wireless Propagation Letters.

[2] S. Velamparambil,et al. Analysis and performance of a distributed memory multilevel fast multipole algorithm , 2005, IEEE Transactions on Antennas and Propagation.

[3] R. Coifman,et al. The fast multipole method for the wave equation: a pedestrian prescription , 1993, IEEE Antennas and Propagation Magazine.

[4] L. Gurel,et al. Efficient Parallelization of the Multilevel Fast Multipole Algorithm for the Solution of Large-Scale Scattering Problems , 2008, IEEE Transactions on Antennas and Propagation.