3D NOC for many-core processors

被引:9
作者
Zia, Aamir [1 ]
Kannan, Sachhidh [1 ]
Chao, H. Jonathan [1 ]
Rose, Garrett S. [1 ]
机构
[1] NYU, Polytech Inst, Metrotech Ctr 5, Brooklyn, NY 11201 USA
关键词
3D integration; CMP; Clos network; Through-silicon vias; Scalability; Low-power; ON-CHIP; PERFORMANCE; TECHNOLOGY; DESIGN; MODEL; ICS;
D O I
10.1016/j.mejo.2011.09.013
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
With an increasing number of processors forming many-core chip multiprocessors (CMP), there exists a need for easily scalable, high-performance and low-power intra-chip communication infrastructure for emerging systems. In CMPs with hundreds of processing elements, 3D integration can be utilized to shorten long wires forming communication links. In this paper, we propose a Clos network-on-chip (CNOC) in conjunction with 3D integration as a viable network topology for many core CMPs. The primary benefit of 3D CNOC is scalability and a clear upper bound on power dissipation. We present the architectural and physical design of 3D CNOC and compare its performance with several other topologies. Comparisons are made among several topologies (fat tree, flattened butterfly, mesh and Clos) showing the power consumption of a 3D CNOC increases only minimally as the network size is scaled from 64 to 512 nodes relative to the other topologies. Furthermore, in a 512-node system, 3D CNOC consumes about 15% less average power than any other topology. We also compare 3D partitioning strategies for these topologies and discuss their effect on wire delay and the number of through-silicon vias. (C) 2011 Elsevier Ltd. All rights reserved.
引用
收藏
页码:1380 / 1390
页数:11
相关论文
共 39 条
  • [11] Dally W., 2003, Principles and Practices of Interconnection Networks
  • [12] Dally WJ, 2001, DES AUT CON, P684, DOI 10.1109/DAC.2001.935594
  • [13] Das R., 2009, DESIGN EVALUATION HI, P175
  • [14] Demystifying 3D ICs: The procs and cons of going vertical
    Davis, WR
    Wilson, J
    Mick, S
    Xu, M
    Hua, H
    Mineo, C
    Sule, AM
    Steer, M
    Franzon, PD
    [J]. IEEE DESIGN & TEST OF COMPUTERS, 2005, 22 (06): : 498 - 510
  • [15] Networks-on-Chip in a Three-Dimensional Environment: A Performance Evaluation
    Feero, Brett Stanley
    Pande, Partha Pratim
    [J]. IEEE TRANSACTIONS ON COMPUTERS, 2009, 58 (01) : 32 - 45
  • [16] RUFT: Simplifying the Fat-tree Topology
    Gomez, C.
    Gilabert, F.
    Gomez, M. E.
    Lopez, P.
    Duato, J.
    [J]. PROCEEDINGS OF THE 2008 14TH IEEE INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED SYSTEMS, 2008, : 153 - 160
  • [17] Grot B, 2009, INT S HIGH PERF COMP, P163, DOI 10.1109/HPCA.2009.4798251
  • [18] Mitigating Memory Wall Effects in High-Clock-Rate and Multicore CMOS 3-D Processor Memory Stacks
    Jacob, Philip
    Zia, Aamir
    Erdogan, Okan
    Belemjian, Paul M.
    Kim, Jin-Woo
    Chu, Michael
    Kraft, Russell P.
    McDonald, John F.
    Bernstein, Kerry
    [J]. PROCEEDINGS OF THE IEEE, 2009, 97 (01) : 108 - 122
  • [19] Jayasimha DN., 2006, On-chip interconnection networks: Why they are different and how to compare them
  • [20] Kahng AB, 2009, DES AUT TEST EUROPE, P423