期刊文献+

一种运算簇间互连通信单元的设计

A Design of Arithmetic Cluster Inter Communication Unit
下载PDF
导出
摘要 在高性能并行处理器设计中,权衡通信效率与硬件设计开销是一个关键的问题。基于此,在基于簇状处理单元的线性阵列处理器架构前提下,提出一种基于多运算簇处理器结构的运算簇间互连通信设计方案,包括通信单元结构和典型数字信号处理数据传输的应用案例分析。实验结果表明,与传统线性阵列处理器结构相比,该方案可使互连通信单元的相应性能提升30%以上。 How to balance the inter communication efficiency and the hardware cost is a key concern for most design of high performance processor. In order to solve this problem, this paper proposes a novel design of arithmetic cluster inter communication structure based on a processor architecture, Multiple Cluster Processor(MCP), on the premise of cluster processing unit of the linear array processor architecture. It mainly includes communication unit structure and the analysis for the communication performance with some typical DSP application. Experimental results show that the design can increase the communication performance about 30% than the other tranditional linear array processor architectures.
出处 《计算机工程》 CAS CSCD 2013年第9期153-156,共4页 Computer Engineering
基金 国家"863"计划基金资助重点项目(2009AA011705)
关键词 互连通信 并行运算 多集群处理器结构 运算簇 线性阵列 通信块 inter communication parallel arithmetic Multiple Cluster Processor(MCP) architecture arithmetic cluster linear array communication block
  • 相关文献

参考文献9

  • 1Kapasi U J, Dally W J, Rixner S, et al. The Imagine Stream Processor[C]//Proc. of IEEE International Conference on Computer Design. [S. 1.]: IEEE Press, 2002: 282-288.
  • 2Malita M, Stefan C: Thi:baut D. Not Multi-, but Many-core: Designing Integral Parallel Architecture for Embedded Computation[C]//Proc. of ACM SIGARCH Computer Architecture News. IS. 1.]: ACM Press, 2007: 32-38.
  • 3Parcerisa J M, Sahuquillo J, Gonz:ilez A, et al. On-chip Interconnects and Instruction Steering Schemes for Clustered Microarchitectures[J]. IEEE Transactions on Parallel and Systems, 2005, 16(2): 130-144.
  • 4刘超.计算机体系结构[M].2版.北京:水利水电出版社,2010.
  • 5Thiebaut D, Malita M. Pipelining the Connex Array[C]//Proc. of BARC'07. Boston, USA: [s. n.], 2007.
  • 6Lee M H, Singh H, Lu Guangming, et al. Design and Implementation of the MorphoSys Reconfigurable Computing Processor[J]. Journal of VLSI Signal Processing Systems, 2000, 24(1): 147-164.
  • 7Chen Wen-Hsiung, Smith C, Fralick S. A Fast Computational Algorithm for the Discrete Cosine Transform[J]. IEEE Transactions on Communications, 1977, 25(9): 1004-1009.
  • 8Calla A M, Stefan G. Matrix Computation on Connex Parallel Architecture[C]//Proc. of International Conference on Signals and Electronic Systems. Gliwice, Poland: Is. n.], 2010: 375- 378.
  • 9Singh H, Lu Guangming, Lee M H. Morphosys: Case Study of a Reconfigurable Computing System Targeting Multimedia Applications[C]//Proc. of IEEE Design Automation Con- ference. IS. 1.]: IEEE Press, 2000: 573-578.

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