[1]NVIDIA Corporation.NVIDIA’s next generation CUDA compute architecture:Fermi[Z].Version 1.1.2009.
[2]NVIDIA Corporation. GeForce GTX 680:The fastest, most efficient GPU ever built[Z].Version 1.0. 2012.
[3]NVIDIA Corporation. NVIDIA’s next generation CUDA compute architecture:Kepler GK110—The fastest,most efficient GPU ever built[Z]. Version 1.0. 2012.
[4]NVIDIA Corporation. NVIDIA CUDA reference manual[Z]. Version 3.2 Beta. 2010.
[5]NVIDIA Corporation. NVIDIA CUDA API reference manual[Z].Version 4.2. 2012.
[6]Cai Jing. GPGPU architecture key technology demonstration and simulator research and extension[D].Changsha:National University of Defense Technology,2009.(in Chinese)
[7]Aammodt M,Bakhoda A,Fung W.Tutorial on GPGPUSim:A performance simulator for massively multithreaded processor research[C]∥Proc of the 42nd Annual IEEE/ACM International Symposium on Microarchitecture,2009:1.
[8]Che Shuai, Boyer M, Meng Jiayuan, et al. A benchmark suite for heterogeneous computing[C]∥Proc of IEEE International Symposium on in Workload Characterization, 2009:1.
[9]NVIDIA Corporation. NVIDIA CUDA SDK3.1 code samples[EB/OL].[20110516].http://docs.nvidia.com/cuda/cudasamples/index.html.
附中文参考文献:
[6]蔡晶. GPGPU体系结构关键技术论证及模拟器研究与扩展[D].长沙:国防科学技术大学,2009.