• DocumentCode
    3074517
  • Title

    Probabilistic Modeling of Frequent Value Bus Encoding Scheme for Low Power Computation

  • Author

    Mehta, Krunal K. ; Kowar, M. ; Sharma, H.R.

  • fYear
    2009
  • fDate
    6-7 March 2009
  • Firstpage
    790
  • Lastpage
    793
  • Abstract
    Exponential increase in scale of integration, is increasing day by day and movement of high volume information among various parts of computing system contributes a lot on overall power budget. Feasibility of system depends on power consumption in modern computing systems. There are various ways to reduce power requirement. Dynamic power requirement is the dominating factor among them. Power consumption due to communication on system-level buses contributes a lot of power consumption. Scheme used to reduce dynamic power in called as bus encoding technique. Various schemes have been proposed in literature to encode data to reduce the number of bus transition. Data volume of information that belongs to computer system is not always used with equal probability, but frequencies of usages are different and high for only few sets of data. Bus encoding scheme suggested with the above fact is called as frequent value encoding (FVE) scheme. This paper provides mathematical model of FVE scheme. Paper aims at providing a framework for evaluation of bus encoding algorithms for 8 to 64-bit information flowing in computing system. Probabilistic model has been drawn and result shows efficiency of FVE scheme.
  • Keywords
    field buses; dynamic power requirement; frequent value bus encoding scheme; power consumption; system-level buses; Capacitance; Costs; Encoding; Energy consumption; Frequency; Hardware; Mathematical model; Power dissipation; Power system modeling; Voltage; Bus Encoding; Bus transition; CMOS; Frequent Value Encoding Scheme; Low power dissipation; VLSI;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Advance Computing Conference, 2009. IACC 2009. IEEE International
  • Conference_Location
    Patiala
  • Print_ISBN
    978-1-4244-2927-1
  • Electronic_ISBN
    978-1-4244-2928-8
  • Type

    conf

  • DOI
    10.1109/IADCC.2009.4809115
  • Filename
    4809115