Developed a CUDA version of the FDTD method and achieved a speedup 40x. Implemented on a NVIDIA Quadro FX 3800 GPU, which has 192 SPs, 1GB global memory, and a memory bandwidth of 51.2 GB/s.
Finite-Difference Time-Domain (FDTD) methods have become a cornerstone in the simulation of electromagnetic phenomena, enabling researchers to solve Maxwell’s equations through discretisation in both ...
Results that may be inaccessible to you are currently showing.
Hide inaccessible results
Feedback