• DocumentCode
    123799
  • Title

    A Service Selection Mechanism Using Fault-Tolerance Techniques

  • Author

    Amario de Souza, Higor ; Pontes Guimaraes, Felipe ; Kon, Fabio ; Macedo Batista, Daniel

  • Author_Institution
    Dept. of Comput. Sci., Univ. of Sao Paulo, Sao Paulo, Brazil
  • fYear
    2014
  • fDate
    5-9 May 2014
  • Firstpage
    214
  • Lastpage
    222
  • Abstract
    Service selection in Service-oriented computing (SOC) is an important step towards obtaining the required level of quality. Fault-Tolerance Techniques (FTT) can be used to enhance the reliability of compositions considering Quality of Service (QoS) requirements. Services that do not perform as expected must be replaced by equivalent services. We propose a mechanism called FTT Service Selector (FSS) to select services that meet QoS constraint values defined by the user. FSS uses a Constraint-Satisfaction Problems (CSP) approach to generate a global optimal service selection. Experiments show that FSS can be used to generate solutions to form compositions with up to ten roles and ten equivalent services. In most cases, FSS can generate solutions in a feasible time.
  • Keywords
    cloud computing; constraint satisfaction problems; quality of service; software fault tolerance; CSP approach; FSS; FTT service selector; QoS constraint values; QoS requirements; SOC; composition reliability; constraint-satisfaction problem approach; fault-tolerance techniques; global optimal service selection; quality of service requirements; service selection mechanism; service-oriented computing; Equations; Fault tolerance; Frequency selective surfaces; Mathematical model; Quality of service; Time factors; compositions; faulttolerance techniques; quality of service; service-oriented computing;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Networks and Distributed Systems (SBRC), 2014 Brazilian Symposium on
  • Conference_Location
    Florianopolis
  • Type

    conf

  • DOI
    10.1109/SBRC.2014.26
  • Filename
    6927138