• DocumentCode
    2637943
  • Title

    High Availability Fault Tolerant Architectures Implemented into FPGAs

  • Author

    Straka, Martin ; Kotasek, Zdenek

  • Author_Institution
    Fac. of Inf. Technol., Brno Univ. of Technol., Brno, Czech Republic
  • fYear
    2009
  • fDate
    27-29 Aug. 2009
  • Firstpage
    108
  • Lastpage
    115
  • Abstract
    In the paper, the methodology of fault tolerant systems design based on field programmable gate array are presented. The architectures are based both on duplex and triple modula redundancy systems to which fault detection capabilities are added, the use of on-line checkers for this purpose is demonstrated. It is described how reliability and availability parameters in triple modula redundancy and duplex structures with checkers can be increased. To demonstrate this, analytical calculations based on Markov reliability model are used. It is also shown how the availability parameters can be affected by the operating environment into which the fault tolerant system is implemented. The principles of generating sequence of fault tolerant architectures with different level of diagnostic are presented.
  • Keywords
    Markov processes; fault tolerance; field programmable gate arrays; FPGA; Markov reliability model; availability parameters; double modula redundancy system; fault detection; fault tolerant architectures; fault tolerant systems; field programmable gate array; on-line checkers; operating environment; reliability parameters; triple modula redundancy system; Availability; Circuit faults; Design methodology; Digital circuits; Digital systems; Fault detection; Fault tolerance; Fault tolerant systems; Field programmable gate arrays; Redundancy; FPGA; TMR; architecture; availability; checker; duplex; fault tolerant system; reliability model;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Digital System Design, Architectures, Methods and Tools, 2009. DSD '09. 12th Euromicro Conference on
  • Conference_Location
    Patras
  • Print_ISBN
    978-0-7695-3782-5
  • Type

    conf

  • DOI
    10.1109/DSD.2009.150
  • Filename
    5350228