• DocumentCode
    585796
  • Title

    “Free” Razor: A novel adaptive voltage scaling low power technique for data path SoC designs

  • Author

    Wu, Yuejian ; Thomson, Sand ; Sun, Han ; Krause, David ; Song Yu ; Kurio, George

  • Author_Institution
    Infinera, Ottawa, ON, Canada
  • fYear
    2012
  • fDate
    12-14 Sept. 2012
  • Firstpage
    33
  • Lastpage
    38
  • Abstract
    This paper proposes a novel adaptive voltage scaling low power design methodology for large System on Chip (SoC) that demands constant data throughput. The proposed technique scales the supply voltage to the SoC based on operating conditions and bit error rate (BER) margin available in a system. It allows occasional timing errors in the circuit and relies on a forward error correction (FEC) that exists in the system to correct the errors. As a result, the proposed technique requires no hardware overhead but yields significant power savings. More importantly, it does not require any circuit modification based on place and route, thus easy to implement and has no impact on time to market. The new technique has been implemented in a complex telecom SoC design and silicon measurement shows power savings up to 46%.
  • Keywords
    error statistics; forward error correction; integrated circuit design; low-power electronics; power aware computing; system-on-chip; BER margin; FEC; adaptive voltage scaling; bit error rate; complex telecom SoC design; data path SoC design; data throughput; forward error correction; low power design methodology; power saving; silicon measurement; supply voltage; system on chip; timing error; Bit error rate; Forward error correction; Power dissipation; Sensors; System-on-a-chip; Timing; Voltage measurement; Low power design; voltage scaling;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    SOC Conference (SOCC), 2012 IEEE International
  • Conference_Location
    Niagara Falls, NY
  • ISSN
    2164-1676
  • Print_ISBN
    978-1-4673-1294-3
  • Type

    conf

  • DOI
    10.1109/SOCC.2012.6398373
  • Filename
    6398373