Acceleration of the DGF-FDTD method on GPU using the CUDA technology

We present a parallel implementation of the discrete Green's function formulation of the finite-difference time-domain (DGF-FDTD) method on a graphics processing unit (GPU). The compute unified device architecture (CUDA) parallel computing platform is applied in the developed implementation. For the sake of example, arrays of Yagi-Uda antennas were simulated with the use of DGF-FDTD on GPU. The efficiency of parallel computations is presented as a function of the number of current elements in such wire antennas. The speedup of the developed implementation approaches maximally 10 relative to the code executed on a multicore central processing unit (CPU). Furthermore, the hybrid technique combining FDTD and its convolution formulation based on DGF was implemented on a heterogeneous CPU-GPU system, facilitating our research on the DGF-FDTD method.