期刊文献+

面向国产CPU的可重构计算系统设计及性能探究 被引量:7

Reconfigurable computing system design and performance exploration towards to domestic CPU
下载PDF
导出
摘要 为了提升国产平台的计算性能,采用国产CPU+FPGA的异构架构,设计了基于国产CPU的可重构计算系统。该系统包括基于国产CPU的主机单元和FPGA可重构加速单元,主机单元负责逻辑判断与管理调度等任务,FPGA负责对计算密集型任务进行加速,并采用OpenCL框架模型进行编程,以缩短FPGA的开发周期。为了验证该系统的性能,采用AES加密算法来测试该系统的计算性能,通过对不同长度的明文进行AES加密测试,并与CPU串行处理结果进行对比,得出:相比于单核FT-1500A CPU串行加密方式,采用可重构计算系统并行加密能够获得120多倍的加速比,且此加速比会随着明文长度的增加而成非线性增大。实验结果表明:基于国产CPU的可重构计算系统能够大幅提升国产平台的计算性能。 In order to improve the performance of domestic computing system,this paper designs a reconfigurable computing system based on domestic CPU.The system consists of host unit based on domestic CPU and FPGA reconfigurable accelerator unit.The CPU acting as the host unit is responsible for logical judgement and task scheduling,while the FPGA is responsible for accelerating the computation task.In order to shorten the FPGA development cycle,the OpenCL framework standard is adopted.To test the peroformance of this system,the AES encryptions with varied length plaintext are tested,compared with the results of CPU serial processing.The system can acquire more than 120 times speedup compared to single FT-1500A CPU core.Experimental results demonstrate that the proposed reconfigurable computing system can improve the computing performance of domestic platform significantly.
作者 彭福来 于治楼 陈乃阔 耿士华 李凯一 PENG Fulai;YU Zhilou;CHEN Naikuo;GENG Shihua;LI Kaiyi(Shandong Chaoyue Digital Control Electronics Co.,Ltd,Shandong Special Computer Key Laboratory,Jinan 250104,China)
出处 《计算机工程与应用》 CSCD 北大核心 2018年第23期36-41,共6页 Computer Engineering and Applications
关键词 可重构计算 国产CPU 现场可编程门阵列(FPGA) AES算法 OPENCL reconfigurable computing system domestic CPU Field-Programmable GateArray(FPGA) AES algorithm OpenCL
  • 相关文献

参考文献5

二级参考文献76

  • 1SchenaM. Microarray Biochip Technology[M]. Eaton Publishing, 2000. 被引量:1
  • 2Friedman N, Linial M, Nachman I, et al. Using Bayesian Networks to Analyze Expression Data[J]. Journal Computational Biology, 2000,7(3-4) : 601-620. 被引量:1
  • 3Lu Y, Lu S, Fotouhi F, et al. Incremental Genetic K-means Algorithm and Its Application in Gene Expression Data Analysis[C]//Proc of BMC Bioinformatics, 2004. 被引量:1
  • 4Ressom H, Wang D, Natarajan P. Adaptive Double Selforganizing Maps for Clustering Gene Expression Profiles [J]. Neural Networks, 2003,16(5/6) : 633-640. 被引量:1
  • 5Su M, Chang H. A New Model of Self-Organizing Neural Networks and Its Application in Data Projection[J]. IEEE Trans on Neural Network,2001,12(1):153-158. 被引量:1
  • 6Gokhale M, Frigo J, Lavenier D. Experience with a Hybrid Processor: K-Means Clustering[J]. The Journal of Supercomputing, 2003,26 (2) : 131-148. 被引量:1
  • 7Lavenier D. FPGA Implementation of the K-means Clustering Algorithm for Hypersprctral Images, 2000. 被引量:1
  • 8Belanovi'c P, Leeser M. A Library of Parameterized Floatingpoint Modules and Their Use[M]. Springer, 2002. 被引量:1
  • 9NVIDIA . CUDA [EB/OL]. [2007-10-08]. http:// www. nvidia.com/cuda. 被引量:1
  • 10AMD. Stream[EB/OL]. [2009-03-12]. http://www. amd. com/ stream. 被引量:1

共引文献27

同被引文献51

引证文献7

二级引证文献10

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