• DocumentCode
    2737884
  • Title

    Energy efficient dual-issue processor for embedded applications

  • Author

    Lozano, Hanni B. ; Ito, Mabo

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Univ. of British Columbia, Vancouver, BC, Canada
  • fYear
    2015
  • fDate
    9-11 April 2015
  • Firstpage
    544
  • Lastpage
    549
  • Abstract
    Low cost and low power scalar embedded processors can issue only a single instruction per cycle which results in longer execution time for applications and consequently lower energy efficiency. Superscalar processor that can issue multiple instructions per cycle are more energy efficient than scalar processors, however, they consume more power which is severely limited in embedded systems that operate on small batteries or energy harvesting power sources. In this paper we propose an energy efficient dual-issue embedded processor that can deliver up to 60% improvement in IPC (instruction-per-cycle) performance with less than 20% increase in power consumption compared to a single-issue embedded processor. In contrast to traditional multi-issue embedded processors that use power intensive superscalar techniques to extract instruction-level parallelism from applications, the proposed processor uses simple hardware techniques to resolve instruction scheduling conflicts. The processor is optimized for implementation on low cost FPGAs suitable for power sensitive embedded applications.
  • Keywords
    microprocessor chips; reduced instruction set computing; IPC; energy efficient dual-issue processor; instruction-level parallelism; instruction-per-cycle performance; low cost scalar embedded processor; low power scalar embedded processor; power consumption; superscalar processor; Automotive engineering; Benchmark testing; Decoding; Field programmable gate arrays; Hardware; Pipelines; Registers; embedded processors; smart devices;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Networking, Sensing and Control (ICNSC), 2015 IEEE 12th International Conference on
  • Conference_Location
    Taipei
  • Type

    conf

  • DOI
    10.1109/ICNSC.2015.7116095
  • Filename
    7116095