Design of a High-Performance CDMA-Based Broadcast-Free Photonic Multi-Core Network on Chip

被引:3
|
作者
Poddar, Soumyajit [1 ,2 ]
Ghosal, Prasun [1 ,2 ]
Rahaman, Hafizur [1 ,2 ]
机构
[1] Indian Inst Engn Sci & Technol, Sibpur, India
[2] Indian Inst Engn Sci & Technol, PO Bot Garden, Howrah 711103, WB, India
关键词
Photonic networks on chip; optical interconnect; code division multiple access; multicore architecture; ON-CHIP;
D O I
10.1145/2839301
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Present-day focus on multicore research has not only increased computing power but also power- and bandwidth-efficient communication among cores. On-chip communication networks have become popular today because of their low energy use and modular structure compared to bus-based interconnects. Silicon photonics has further boosted the performance of on-chip interconnection networks with its low energy-delay product and high reliability. In current multicore Network-on-Chip (NoC) architectures, photonics is playing an important role in transferring large volumes of data both on- and off-chip. The problem addressed in this work is the issue of broadcast traffic arising due to invalidation requests from on-chip cache memories. Although such traffic is typically less than 1% of total traffic, it can easily present a high load on network resources, creating congestion and degrading performance. In this article, we propose a CDMA-based, secure, scalable, and energy-efficient technique to eliminate broadcast invalidations and increase overall performance. Experimental results indicate a performance boost up to 22.2% over a competing Photonic NoC and up to 57.4% over Electrical Mesh-based NoC when the proposed technique is used. Additional hardware deployed has an area overhead of less than 1%, whereas total energy consumed is at par with other state-of-the-art techniques.
引用
收藏
页数:30
相关论文
共 50 条
  • [41] A function-based on-chip communication design in the heterogeneous multi-core architecture
    Chen, Tianzhou
    Chen, Guobing
    Dai, Hongjun
    Shi, Qinsong
    MUE: 2007 INTERNATIONAL CONFERENCE ON MULTIMEDIA AND UBIQUITOUS ENGINEERING, PROCEEDINGS, 2007, : 1086 - +
  • [42] LumiNOC: A Power-Efficient, High-Performance, Photonic Network-on-Chip
    Li, Cheng
    Browning, Mark
    Gratz, Paul V.
    Palermo, Samuel
    IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, 2014, 33 (06) : 826 - 838
  • [43] Compaction-free Compressed Cache for High Performance Multi-core System
    Hsu, Po-Yang
    Lin, Pei-Lan
    Hwang, TingTing
    2014 IEEE/ACM INTERNATIONAL CONFERENCE ON COMPUTER-AIDED DESIGN (ICCAD), 2014, : 140 - 147
  • [44] Realization and Performance Comparison of Sequential and Weak Memory Consistency Models in Network-on-Chip based Multi-core Systems
    Naeem, Abdul
    Chen, Xiaowen
    Lu, Zhonghai
    Jantsch, Axel
    2011 16TH ASIA AND SOUTH PACIFIC DESIGN AUTOMATION CONFERENCE (ASP-DAC), 2011,
  • [45] Hybrid silicon-photonic network-on-chip for future generations of high-performance many-core systems
    Achraf Ben Ahmed
    Abderazek Ben Abdallah
    The Journal of Supercomputing, 2015, 71 : 4446 - 4475
  • [46] Microring fault-resilient photonic network-on-chip for reliable high-performance many-core systems
    Michael Meyer
    Yuichi Okuyama
    Abderazek Ben Abdallah
    The Journal of Supercomputing, 2017, 73 : 1567 - 1599
  • [47] Microring fault-resilient photonic network-on-chip for reliable high-performance many-core systems
    Meyer, Michael
    Okuyama, Yuichi
    Abdallah, Abderazek Ben
    JOURNAL OF SUPERCOMPUTING, 2017, 73 (04): : 1567 - 1599
  • [48] Hybrid silicon-photonic network-on-chip for future generations of high-performance many-core systems
    Ben Ahmed, Achraf
    Ben Abdallah, Abderazek
    JOURNAL OF SUPERCOMPUTING, 2015, 71 (12): : 4446 - 4475
  • [49] Design of High-Performance and General-Purpose Satellite Management Unit Based on Rad-Hard Multi-Core SoCand Linux
    Li, Lu
    He, Junwang
    Xu, Dongxiao
    Chen, Wen
    Yu, Jinpei
    Li, Huawang
    AEROSPACE, 2023, 10 (02)
  • [50] CapEx-minimized incremental network design based on multi-core fibers
    Li Y.
    Hua N.
    Zheng X.
    Qinghua Daxue Xuebao/Journal of Tsinghua University, 2016, 56 (09): : 937 - 941