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 50 条
  • [41] Approximate Set Similarity Join Using Many-Core Processors
    Sugano, Kenta
    Amagasa, Toshiyuki
    Kitagawa, Hiroyuki
    DATABASE AND EXPERT SYSTEMS APPLICATIONS (DEXA 2018), PT II, 2018, 11030 : 214 - 222
  • [42] Tailoring Genetic Algorithm for Resource Scheduling in Many-Core Processors
    Hu, Xiande
    Li, Jingming
    Cheng, Jiaxing
    PROCEEDINGS OF THE 2015 2ND INTERNATIONAL CONFERENCE ON ELECTRICAL, COMPUTER ENGINEERING AND ELECTRONICS (ICECEE 2015), 2015, 24 : 465 - 471
  • [43] Reducing the burden of parallel loop schedulers for many-core processors
    Arif, Mahwish
    Vandierendonck, Hans
    CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2021, 33 (13):
  • [44] Algorithmic Patterns for H-Matrices on Many-Core Processors
    Zaspel, Peter
    JOURNAL OF SCIENTIFIC COMPUTING, 2019, 78 (02) : 1174 - 1206
  • [45] All-pairs computations on many-core graphics processors
    Sarje, Abhinav
    Aluru, Srinivas
    PARALLEL COMPUTING, 2013, 39 (02) : 79 - 93
  • [46] Method of independent debug system design for many-core processors
    Zhang, Ming
    Shi, Wei
    Guo, Yu-Feng
    Zhang, Min-Xuan
    Hunan Daxue Xuebao/Journal of Hunan University Natural Sciences, 2015, 42 (04): : 93 - 99
  • [47] A Memory-Efficient NoC System for OpenCL Many-core Platform
    Yen, Chien-Hsuan
    Chen, Chung-Ho
    Chen, Kuan-Chung
    2015 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS), 2015, : 1386 - 1389
  • [48] System management recovery in NoC-based many-core systems
    Vinicius Fochi
    Luciano L. Caimi
    Marcelo H. da Silva
    Fernando Gehm Moraes
    Analog Integrated Circuits and Signal Processing, 2021, 106 : 85 - 98
  • [49] Adaptive fault-tolerant architecture and routing algorithm for reliable many-core 3D-NoC systems
    Ben Ahmed, Akram
    Ben Abdallah, Abderazek
    JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 2016, 93-94 : 30 - 43
  • [50] Thermal Investigation Into Power Multiplexing for Homogeneous Many-Core Processors
    Gupta, Man Prakash
    Cho, Minki
    Mukhopadhyay, Saibal
    Kumar, Satish
    JOURNAL OF HEAT TRANSFER-TRANSACTIONS OF THE ASME, 2012, 134 (06):