期刊文献+

基于OpenCL的并行FFT算法的设计研究

下载PDF
导出
摘要 本设计以快速傅里叶变换(FFT)为研究对象,对FFT算法的并行化进行解析,分解并行指令,并且给编写的Open CL内核做了基本的线程优化,在Altera So C上实现加速算法仿真,并与非加速情况进行比较。实验结果表明,与采用coretex-A9串行算法相比,本文设计的基于FPGA平台的Open CL的并行算法在FFT-16,FFT-32分别可在时间上分别节省32.9%,46.81%,由此可见,随着FFT的点数的增大,时间节省的比例也越来越高。
作者 徐庚辉
出处 《黑龙江科技信息》 2016年第29期173-173,共1页 Heilongjiang Science and Technology Information
  • 相关文献

参考文献2

二级参考文献38

  • 1陈国良,梁维发,沈鸿.并行图论算法研究进展[J].计算机研究与发展,1995,32(9):1-16. 被引量:13
  • 2Chen G L, Sun G Z, Zhang Y Q, et al. Study on parallel computing. J Comput Sci Tech, 2006.21(5): 665--673.
  • 3Grama A, Gupta A, Karypis G, et al. Introduction to parallel computing. Boston: Benjaming/Cummings Publish Company, Inc., 2003.
  • 4Chen G L. A partitioning selection algorithm on multiprocessors. J Comput Sci Tech, 1988, 3(4): 241--250.
  • 5Zhang F, Chen G L, Zhang Z Q. OpenMP on Networks of Workstations for Software DSMs. J Comput Sci Tech, 2002, 17(1): 90--100.
  • 6Sutter H, Larus J. Software and the concurrency revolution. Q focus: Multiprocessors, 2005, 3(7): 54--62.
  • 7Rajkumar B, Chee S Y, Srikumar V. Market-oriented cloud computing: Vision, hype, and reality for delivering IT services as computing utilities. In: Proceedings of the 10th IEEE International Conference on High Performance Computing and Communications, 2008 Sept 25-27, Dalian. Los Alamitos, CA: IEEE CS Press, 2008. 15--22.
  • 8Asanovic K, Bodik R, James J, et al. The landscape of parallel computing research: A view from Berkeley. Technical Report, Electrical Engineering and Computer Sciences, University of California, Berkeley. 2006.
  • 9Zhang Y Q, Chen G L, Sun G Z, et al. Models of parallel computation: a survey and classification. Front Comput Sci China, 2007, 1(2): 156--165.
  • 10Sun X H. Scalable computing in the multicore era. In: Proceedings of the Inaugural Symposium on Parallel Algorithms, Architechures and Programming, 2008 8ep 16-18, Hefei. Hefei: University of Science and Technology of China Press, 2008. 1--18.

共引文献117

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部