High Performance Computing of Complex Electromagnetic Algorithms Based on GPU/CPU Heterogeneous Platform and Its Applications to EM Scattering and Multilayered Medium Structure

Table 1

Detailed parameters of GPU platform.

Parameters	Fermi; GF104	Kepler; GK104	Kepler; GK110
Compute ability	2.1	3.0	3.5
Threads/warp	32	32	32
Max warps/multiprocessor	48	64	64
Max threads/multiprocessor	1536	2048	2048
Max thread Blocks/Multiprocessor	8	16	16
32-bit registers/multiprocessor	32768	65536	65536
Max registers/thread	63	63	255
Max threads/thread block	1024	1024	1024
Max X-grid dimension		2^32−1	2^32−1
Dynamic parallelism	×	×	√
Used in numerical algorithms	—	CG-FFT	FMM; ACA