期刊文献+

分级环片上网络互连 被引量:5

A Kind of Hierarchical Ring Interconnection Networks-on-Chip
下载PDF
导出
摘要 在大规模、超大规模片上互连网络中,因为二维互连方式的性能较差而使多维互连方式成为可选方案之一.文中首先基于区域划分设计了一种分级环互连结构,分析了其静态互连特性,然后基于卡诺图编码设计了一种分级环互连的路由结构以及寻径方法,在均匀通信模式测试了不同的分级环级联链路缓冲区设置方法下网络的性能,详细分析了按照等比序列设置分级环级联链路缓冲区时分级环互连方式的动态网络特性,最后根据互连性能与Mesh等二维片上互连方式比较的结果,给出了分级环互连方式的使用场合.实验结果表明,虽然在较小规模网络中性能较差,但是分级环互连方式能以较低的成本、较高的性能实现大规模、超大规模片上网络的互连,其中单环分级互连方式在较低网络负载下综合性能更好,而双环分级互连方式则具有更大的网络负载能力,在较高网络负载下性能更好. and ver achieve Networ t The interconnection performance of the 2-D Networks-on-Chip is so poor in the large y large scales that the multi-dimensional interconnections become one of the choices to better interconnection performance This paper gives a design of a kind of hierarchical ring ks-on-Chip based on the regional partition and analyses the static characteristics of the hi- erarchical rings. Then it proposes a kind of the hierarchical ring router and the routing algorithm based on the Karnaugh map coding. It tests the characteristics with different sets of the buffers in the hierarchical router nodes under the uniform random communication pattern and gives a de- tailed analysis of the dynamic characteristics of the hierarchical rings with the buffers in the hier- archical router nodes set in accordance with the geometric sequence under the global uniform ran- dom communication pattern. Finally, it points out the cases when the hierarchical rings work well according to the comparison of the characteristics of the hierarchical ring intereonnections and some of the 2-D Network-on-Chip interconnections such as Mesh. The results show that, al- though working poor in small scales, the hierarchical rings can interconnect the large and very large scales Networks-on-Chip with lower cost and higher performance. Comparatively, the single-hierarchical ring interconnection works better under lower loads and the double-hierarchical ring interconnection has larger load capability and works better under higher loads.
出处 《计算机学报》 EI CSCD 北大核心 2010年第2期326-334,共9页 Chinese Journal of Computers
基金 国家自然科学基金(60773149) 国家"八六三"高技术研究发展计划项目基金(2006AA01A101 2008AA01Z108) 国家"九七三"重点基础研究发展规划项目基金(2007CB310900)资助
关键词 片上多处理器 片上网络 分级环 性能分析 chip multiprocessor networks-on-chip hierarchical ring performance analysis
  • 相关文献

参考文献13

  • 1王炜,汤志忠,乔林.片上多处理器互连技术综述[J].计算机科学,2008,35(9):7-8. 被引量:7
  • 2Pande P P, Jones M, Ivanov A, Saleh R. Performance evaluation and design trade-offs for network-on-chip Interconnect architectures. IEEE Transactions on Computers, 2005, 54 (8) : 1025-1040.
  • 3王炜,乔林,杨广文,汤志忠.片上二维网络互连性能分析[J].计算机研究与发展,2009,46(10):1601-1611. 被引量:8
  • 4王炜,乔林,杨广文,汤志忠.扩展二维网格片上互连性能分析[J].清华大学学报(自然科学版),2010,50(1):161-164. 被引量:5
  • 5郑纬民,汤志忠,计算机系统结构.第2版.北京:清华大学出版社,2006.
  • 6Ahmad F, Yuan Xin. Communication characteristics in the NAS parallel benchmarks//Proeeedings of the International Conference on Parallel and Distributed Computing Systems (PDCS' 02). Cambridge, USA, 2002. IASTED/ACTA Press, 2002:724-729.
  • 7Vetter J S, Mueller F. Communication characteristics of large-scale scientific applications for contemporary cluster architectures. Journal of Parallel and Distributed Computing, 2003, 63(9): 853-865.
  • 8Kim JunSeong, Lilja D J. Characterization of communication patterns in message-passing parallel scientific application programs//Proceedings of the Network-Based Parallel Computing: Communication, Architecture, and Applications (CANPC'98). Las Vegas, Nevada, USA, 1998:202-216.
  • 9Manjikian N. Prototyping a hierarchical ring interconnect for system on-chip multiprocessor//Proceedings of the 2nd Annual IEEE Northeast Workshop on Circuits and Systems (NEWCAS 2004), 2004:85-88.
  • 10Denneau M M. The Yorktown simulation engine/ /Proceedings of the 19th Design Automation Conference. Las Vegas, Nevada, 1982:55-59.

二级参考文献32

  • 1Partha Pratim Pande, Michael Joncs, Andre Ivanov, et al. Performance evaluation and design trade-offs for network-on- chip interconnect architectures [J]. IEEE Trans on Computers, 2005, 54(8): 1025-1040.
  • 2John Kim, James Balfour, William J Dally. Flattened butterfly topology for on-chip networks [C]//Proc of the 40th IEEE/ACM Int Symp on Microarchitecture (MICRO- 40). Los Alamitos, CA: IEEE Computer Society, 2007: 172-182.
  • 3Strauss Karin, Shen Xiaowei, Torrellas Josep. Uncorq: Unconstrained snoop request delivery in embedded-ring muhiproeessors [C]//Proc of the 40th IEEE/ACM Int Symp on Microarchitecture (MICRO-40). Los Alamitos, CA: IEEE Computer Society, 2007 : 327-339.
  • 4Chang M Frank, Cong Jason, Kaplan Adam, et al. CMP network on-chip overlaid with multi-band RF-intereonnect [C]//Proc of the 14th Int Symp on High-Performance Computer Architecture (HPCA'08). Los Alamitos, CA: IEEE Computer Society, 2008:191-202.
  • 5Paul. Gratz, Boris Grot, Stephen W Keckler. Regional congestion awareness for load balance in networks-on-chip [C]//Proc of the 14th Int Symp on High-Performance Computer Architecture ( HPCA'08 ). Los Alamitos, CA: IEEE Computer Society, 2008:203-214.
  • 6Reetuparna Das, Asit K Mishra, Chrysostomos Nicopoulos, et al. Performance and power optimization through data compression in network-on-chip architectures [C] //Proc of the 14th Int Symp High-Performance Computer Architccture (HPCA'08). Los Alamitos, CA: IEEE Computer Society. 2008 : 215-225.
  • 7John Kim, William J Dally, Steve Scott, et al. Technology driven, highly-scalable dragonfly topology[C]//Proc of the 35th Int Symp on Computer Architecture (ISCA'08), Los Alamitos, CA: IEEE Computer Society, 2008:77-88.
  • 8Lee Jae W, Man Cheuk Ng, Asanovic Krste. Globally synchronized frames for guaranteed quality-of-service in on- chip networks [C] //Proc of the 35th Int Symp on Computer Architecture ( 1SCA'08 ). Los Alamitos, CA:IEEE Computer Society, 2008:89-100.
  • 9Martha Merealdi Kim, John D Davis, Mark Oskin, et al. Polymorphic on-chip networks [C]//Proc of the 35th Int Symp on Computer Architecture (ISCA'08). Los Alamitos, CA: IEEE Computer Society, 2008:101-112.
  • 10Dana Vantrease, Robert Schrciber, Matteo Monchiero, et al. Corona: system implications of emcrging nanophotonic technology [C] //Proc of the 35th Int Symp on Computer Architecture ( ISCA'08 ). Los Alamitos, CA: IEEE Computer Society, 2008:153-164.

共引文献12

同被引文献107

  • 1徐罗娜,刘三阳,孙玉涛.Star网络的限制边连通度[J].山东理工大学学报(自然科学版),2007,21(3):12-14. 被引量:3
  • 2Dally W J, Towles B. Route packets, not wires:on-chip intereonnection networks[C] // Proceedings of the 38th Design Automation Conference (DAC 2001). New York, NY: ACM, 2001 :684- 689.
  • 3Salminen E, Kulmala A, Hamalainen T D. On network-on-chip comparison[C] // Proceedings of the Tenth Euromiero Conference on Digital System Design: Architectures, Methods and Tools (DSD 2007). Los Alamitos,CA.. IEEE Computer Society, 2007: 503-510.
  • 4Neeb C, Wehn N. Designing efficient irregular networks for heterogeneous Systems-on-Chip [C] // Proceedings of the Ninth Euromicro Conference on Digital System Design: Architectures, Methods and Tools (DSD 2006). Los Alamitos, CA.. IEEE Computer Society, 2006 : 665-672.
  • 5Salminen E, Kangas T, Hamalainen T D, et al. HIBI Communication Network for System-on-Chip[J]. The Journal of VLSI Signal Processing, 2006,43(2/3) : 185-205.
  • 6Xu Junming. Topological Structure and Analysis of Interconnection Networks[M]. Dordrecht/ Boston/London: Kluwer Academic Publishers, 2001.
  • 7Ainsworth T W,Pinkston T M. Characterizing the Cell EIB onchip network[J]. IEEE Micro,2007,27(5)..6-14.
  • 8Karim F, Nguyen A, Dey S, et al. On-chip communication architecture for OC-768 network processors[C]//Proceedings of the 38th Design Automation Conference (DAC 2001). New York, NY: ACM, 2001 : 678-683.
  • 9Coppola M, Loeatelli R, Maruccia G, et al. Spidergon.. a novel onchip communication network[C]//Proceedings of the 2004 International Symposium on System on Chip. Piscataway, NJ: IEEE, 2004 : 15.
  • 10Liu Youyao, Han Jungang, Du HuimirL DL(2m) : A new scalable interconnection network for system-on-chip[J]. Journal of Computesrs, 2009,4 (3) : 201-207.

引证文献5

二级引证文献19

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