期刊文献+

龙芯2E多处理器芯片组的设计与实现

Design and implementation of multi-processor chipset based on Godson 2E CPU
下载PDF
导出
摘要 提出了一种面向高性能计算机的多处理器芯片组的设计,其主要特点是支持多处理器通过芯片组和交换芯片两级互连,全局地址空间和多处理器同步支持。给出了芯片组的组成结构、设计原则和关键技术,设计并实现了基于龙芯2E处理器的多处理器芯片组。目前,已采用FPGA平台对该芯片组进行验证和测试,以该芯片组为核心的四处理器原型系统完成B IOS引导和操作系统运行,经过实测处理器的访问请求通过芯片组延迟小于0.5μs,芯片组内处理器通信带宽达到500 Mbps。 This paper introduced the design of a multi-processor chipset for high performance computer, whose features were the two-layer interconnection by chipset and router, global address space (GAS) and the support for synchronization. It described the architecture, design principles and key techniques of the multi-processor chipset ,and the FPGA implementation of the chipset based on Godson 2E CPU was presented. The 4-CPU prototype system could run Linux operating system. The expe -riment shows that the latency across the chipset is less than 0. 5 μs and the bandwidth between two CPUs is up to 500 Mbps.
出处 《计算机应用研究》 CSCD 北大核心 2008年第5期1465-1469,1473,共6页 Application Research of Computers
基金 中国科学院创新课题资助项目(20054010) 中国科学院计算技术研究所创新课题资助项目(20046080)
关键词 多处理器 芯片组 全局地址空间 龙芯2E处理器 multi-processor chipset global address space (GAS) Godson 2E CPU
  • 相关文献

参考文献14

  • 1CULLER D,SINGH J P, GUPTA A. Parallel computer architecture:a hardware/software approach[ n]. [ S. l. ] : Morgan Kaufmann Press, 1998.
  • 2TOURANCHEAU B, WESTRELIN R. Support for MPI at the network interface level [ C ]//Advances in Parallel Virtual Machine and Message Passing Interface, LNCS 2131. [ S. l. ] :Springer,2001.
  • 3LIU J, CHANDRASEKARAN B, WU J ,et al. Performance comparison of MPI implementations over lnfiniBand, Myrinet and Quadrics[ C]// Proc of International Conference for High Performance Computing and Communications. 2003.
  • 4BRIGHTWELL R, PEDRETTI K, UNDERWOOD K D. Initial performance evaluation of the cray SeaStar interconnect[ C]//Proc of the 13th Symposium on High Performance Interconnects. 2005.
  • 5BRIGHTWELL R. A new MPI implementation for cray SHMEM [ C ]//Advances in Parallel Virtual Machine and Message Passing Interface, LNCS 3241. [S. l. ] :Springer,2004.
  • 6BRIGHTWELL R. A comparison of three MPI implementations for red storm[ C ]//Advances in Parallel Virtual Machine and Message Passing Interface, LNCS 3666. [ S. l. ] :Springer,2005.
  • 7ALMASI G, BELLOFATTO R, BRUNHEROTO J, et al. An overview of the blue gene-L system software organization[ C]//Proc of the 9th International Euro-Par Conference on Parallel Processing. Klagenfurt, Austria : [ s. n. ] , 2003.
  • 8LEE M,DALLY J E,FARJAD-RAD W J,et al. CMOS high-speed I/ Os-present and future [ C ]//Proc of the 21st International Conference on Computer Design. 2003.
  • 9HOROWITZ M,YANG C K K, SIDIROPOULOS S. High speed electrical signaling: overview and limitation[ J]. IEEE MICRO, 1998, 18(1) :12-24.
  • 10胡伟武,张福新,李祖松.龙芯2号处理器设计和性能分析[J].计算机研究与发展,2006,43(6):959-966. 被引量:37

二级参考文献11

  • 1MIPS Ⅳ instruction set. http://www.mips.com, 1995
  • 2Divid Patterson, John Hennessy. Computer A rchitecture: AQuantitative Approach. San Francisco: Morgan Kaufmann, 1996
  • 3R. Kessler. The Alpha 21264 microprocessor, IEEE Micro,1999, 19(2): 24-36
  • 4Kenneth Yeager. The MIPS R10000 superscalar microprocessor.IEEE Micro, 1996, 16(3): 28-41
  • 5Tim Horel, Gary Lauterbach. UntraSparc-Ⅲ: Designing third-generation 64-bit performance. IEEE Micro, 1999, 19 (3) : 73-85
  • 6Ashok Kumar, The HP PA 8000 RISC CPU. IEEE Micro,1997, 17(2): 27-32
  • 7J. Tendler, S. Dodson, S. Fields, et al. Power 4 system microarchitecture. New York: IBM Corporation, 2001
  • 8G. Hinton, D, Sager, M. Upton, et al. The microarchitecture of the Pentium Ⅳ processor. Intel Technology Journal, 2001.http://www.intel. com/technology/iti/q12001/pdf/art_2.pdf
  • 9J. Huck, et al.Introducing the IA-64 architecture. IEEE Micro,2000, 20(5): 12-23
  • 10J. Henning. SPEC CPU2000: Measuring CPU performance in the new millennium. Computer, 2000, 33(7): 28-35

共引文献36

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