期刊文献+

FPGA异构计算平台及其应用 被引量:11

FPGA Heterogeneous Computing Platform and Its Applications
下载PDF
导出
摘要 当前摩尔定律已遇到瓶颈,通用处理器的计算能力受到制约,而深度学习等新型企业计算对数据中心服务器功耗和计算性能提出了更大的挑战。现场可编程门阵列(FieldProgrammable Gate Array,FPGA)具有高性能、低功耗等特点,FPGA异构计算平台可有效解决数据中心能效问题。文章阐述了FPGA异构计算平台的硬件体系结构和软件编程模型,分析了基于Open CL的FPGA异构计算平台的高性能、低功耗、动态可重构等优势。应用实践表明,FPGA异构平台可在降低系统功耗的同时提升系统性能,从而实现系统能效的有效提升。 Moore's law encounters a bottleneck today. Computing power of the general purpose processor is restricted. At the same time, new types of enterprise computing such as deep learning bring more challenges to the power consumption and computational performance of the data center. FPGA has the advantages of high performance and low power consumption. FPGA heterogeneous platform can effectively solve the energy efficiency problems of the data center. In this paper, the hardware architecture and software programming model of FPGA heterogeneous platform are introduced first. Then advantages of this platform are analyzed. The applications of FPGA heterogeneous platform show that it can improve the system performance while reducing the system power consumption, so as to achieve the effective improvement of the system energy efficiency.
出处 《电力信息与通信技术》 2016年第7期6-11,共6页 Electric Power Information and Communication Technology
基金 国家863高技术研究发展计划项目(2015AA015301)
关键词 FPGA OPENCL 异构计算 可重构 FPGA OpenCL heterogeneous computing reconfigurable
  • 相关文献

参考文献5

二级参考文献49

  • 1吴恩华.图形处理器用于通用计算的技术、现状及其挑战[J].软件学报,2004,15(10):1493-1504. 被引量:141
  • 2NVIDIA . CUDA [EB/OL]. [2007-10-08]. http:// www. nvidia.com/cuda.
  • 3AMD. Stream[EB/OL]. [2009-03-12]. http://www. amd. com/ stream.
  • 4Khronos Group. The OpenCL specification [EB/OL]. [2010-09-20]. http: //www. khronos. org/openel/.
  • 5Jens Breitbart, Claudia Fohry. OpenCL-an effective program ruing model for data parallel computations at the cell broadband engine [C]. Los Alamitos: IEEE Computer Society Press, 2010.
  • 6John E Stone, David Gohara, Guochun Shi. OpenCL: A parallel programming standard for heterogeneous computing systems [C]. Los Alamitos: IEEE Computer Society Press, 2010: 66-73.
  • 7Martin Jurecko, Jana Kocisova. Evaluation framework for GPU performance based on OpenCL standard [C]. Los Alamitos: IEEE Computer Soeiety Press, 2010: 256-261.
  • 8Microsoft. DirectCompute [EB/OL]. http://msdn. microsoft.com/zh-cn/directx, 2010.
  • 9Ryo Aoki, Shuiehi Oikawa, Ryo Tshehiyama, et al. Hybrid OpenCL: Connecting different OpenCL implementations over network [C]. Los Alamitos: IEEE Computer Society Press, 2010: 2729-2735.
  • 10Amnon Barak, Tal Ben-Num, Ely Levy, et al. A package for OpenCL based heterogeneous computing on clusters with many GPU devices [C]. Los Alamitos: IEEE Computer Society Press, 2010.

共引文献163

同被引文献62

引证文献11

二级引证文献42

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