期刊文献+

一种多核微处理器互连接口的设计与性能分析 被引量:3

Design and Performance Analysis of an Interconnect Interface for Multi-Core Microprocessor
下载PDF
导出
摘要 并行是提高计算机性能最主要的方法,随着集成电路生产工艺的不断发展,除了在单个芯片内集成更多的处理器核外,通过集成高速互连网络接口构建多路并行系统一直是提高高性能计算机并行性的主要方式。提出了一种面向多核微处理器的互连接口的设计方案,基于精简的PCI-E总线协议,采用高速串行数据传输技术,支持Cache一致性报文和大块数据传输报文,能够用于实现4个处理器的直接互连。模拟结果表明,优化设计的互连接口每个接口能够实现64Gbps的双向最大有效带宽,最小传输延迟为120ns,能够较好平衡不同报文类型对带宽和传输延时的要求。 Parallelism is the most important way to improve the performance of computer. With the development of the integrated circuits' manufacture process,besides integrating more processor cores into one processor chip,building multi-way parallelism system through high-speed interconnect interface is the main method to increase the parallelism of high-performance computer. A design scheme of an interconnect interface for multi-core microprocessor was proposed. The proposed interface was based on a simplified PCI Express bus protocol and adopted the technology of high-speed serial data transferring. Cache coherence packet and large block data transfer packet were all supported. The interface can be used for connecting four processor nodes directly. Simulation results show that the max valid bandwidth per interface can reach 64Gbps and the minimum transfer delay is 120ns. The balance of the bandwidth and the transfer delay is reached,meeting the requirement of transferring different type of packets.
出处 《国防科技大学学报》 EI CAS CSCD 北大核心 2010年第4期94-99,共6页 Journal of National University of Defense Technology
基金 国家"863"计划项目(2009AA01Z124)
关键词 多核处理器 互连 PCI-E multi-core processor interconnect PCI-E
  • 相关文献

参考文献8

  • 1Intel Inc,320412-001US.An Introduction to the Intel QuickPath Interconnect[S].Intel Inc,2009:3-21.
  • 2The HyperTransport Consortium,White Paper HTC-WP02.HyperTransport I/O Technology Overview-An Optimized Low-latency Board-level Architecture[S].The HyperTransport Consortium,2004:3-21.
  • 3Ravi B,Don A,Tom S.PCI Express System Architecture[M].USA:MindShare Inc,2003.
  • 4张峰.高速信号传输技术综述.信息技术快报,2008,6(2):1-5.
  • 5Cadence Inc.I-IPA01-0140-USR Rev 14.10 Gigabit Ethernet XAUI/XGXS Physical Coding Sublayer User Guide[S].Cadence Inc,2007:6-32.
  • 6刘涛,刘光明,郭御风.一种高性能I/O互连协议链路接口的实现[J].计算机工程与科学,2006,28(4):130-132. 被引量:2
  • 7David S,Alexander G,Ulrich B.A Versatile,Low Latency HyperTransport Core[C] //Proceedings of the ACM/SIGDA 15th International Symposium on Field Programmable Gate Arrays,2007:45-52.
  • 8田颖瑞,周宏伟.微处理器直连接口链路层重传机制的设计与验证[D].长沙:国防科技大学,2009.

二级参考文献4

  • 1Jar Trodden, Don Anderson.p.cm ,HyperTransport System Architecture[M].Mindshare Inc,2003.
  • 2HyperTransport. I/O Link Specification (Revision 1.05)[Z].2003.
  • 3Xilinx HyperTransport Lite Interface参考设计[EB/OL].http://www.xilinx.com,2002-05.
  • 4Virtex-Ⅱ Platform FPGA Handbook[EB/OL].http://www.xilinx.com,2002-05.

共引文献1

同被引文献10

  • 1刘涛,刘光明,郭御风.一种高性能I/O互连协议链路接口的实现[J].计算机工程与科学,2006,28(4):130-132. 被引量:2
  • 2Cavalli M. Advanced Interconnect Standards Blend Serial andParallel Techniques for Best Performance and Scalability [ S ]. USA. HyperTransport Consortium. 2007 : 1 - 2. HTC_WI:02-2304.
  • 3HyperTnmsport I/O Technology Overview[ S]. USA. HyperTransport Consortium. 2004.
  • 4Holden B. Latency Comparison between HyperTransport and PCI Express in Communications Systems [ S ]. USA. HyperTransport Consortium. 2007.
  • 5320412 -001US. An Introduetion to the Intel QuickPath Intereanneet[ S]. Intel Ine, 2009:3 -21.
  • 6The HyperTransport Consortium. HyperTransport I/O Link Specification Revision 3.00d [ S ]. USA, HyperTransport Consortium. 2008:41 - 44, 49 - 50.
  • 7陈超.Hyper Transport超传输关键技术研究[D].长沙:国防科学技术大学,2009.
  • 8Ravi B, Don A, Tom S. PCI Express System Architecture[M]. USA : MindShare. Inc, 2003.
  • 9王云霏,王飙,李媛,孙战先.一种应用于多路直连CMP的混合一致性协议[J].计算机工程,2017,34(7):38-43. 被引量:1
  • 10王锦涵,李研,班冬松,陶涛.基于国产处理器直连接口的设计与验证[J].计算机与数字工程,2019,47(11):2700-2704. 被引量:3

引证文献3

二级引证文献9

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