• DocumentCode
    726401
  • Title

    OSFA: A new paradigm of gate-sizing for power/performance optimizations under multiple operating conditions

  • Author

    Roy, Subhendu ; Derong Liu ; Junhyung Um ; Pan, David Z.

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Univ. of Texas at Austin, Austin, TX, USA
  • fYear
    2015
  • fDate
    8-12 June 2015
  • Firstpage
    1
  • Lastpage
    6
  • Abstract
    Modern SoCs and microprocessors, e.g., those in smart phones and laptops, typically have multiple operating conditions, such as video streaming, web browsing, standby, and so on. They will have different performance targets and run under different supply voltages. Gate sizing (with threshold voltage assignment) is a fundamental step for power/performance optimization. However, conventional gate sizing algorithms only consider one scenario, e.g., the performance-critical operating condition, which may be over-design for other operating conditions. In this paper, we present a new paradigm of gate sizing, OSFA (One-Size-Fits-All), which performs power/performance optimizations across multiple operating conditions. Based on OSFA, we also adjust the supply voltage targeting overall power optimization. Experimental results on industry-strength benchmarks demonstrate that compared with conventional approach OSFA could provide an average 6.1% reduction in power without performance loss.
  • Keywords
    microprocessor chips; power aware computing; system-on-chip; OSFA paradigm; SoC; gate sizing algorithms; gate-sizing paradigm; microprocessors; one-size-fits-all paradigm; performance targets; performance-critical operating condition; power-performance optimization; supply voltages; system-on-chip; threshold voltage assignment; Delays; Libraries; Logic gates; Optimization; Sensitivity; Switches;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Design Automation Conference (DAC), 2015 52nd ACM/EDAC/IEEE
  • Conference_Location
    San Francisco, CA
  • Type

    conf

  • DOI
    10.1145/2744769.2744885
  • Filename
    7167314