• DocumentCode
    3154760
  • Title

    A Model Driven Architecture approach to fault tolerance in Service Oriented Architectures, a performance study

  • Author

    Alodib, Mohammed ; Bordbar, Behzad

  • Author_Institution
    Sch. of Comput. Sci., Univ. of Birmingham, Birmingham
  • fYear
    2008
  • fDate
    16-16 Sept. 2008
  • Firstpage
    293
  • Lastpage
    300
  • Abstract
    In modern service oriented architectures (SoA) identifying the occurrences of failure is a crucial task, which can be carried out by the creation of diagnosers to monitor the behavior of the system. Model driven architecture (MDA) can be used to automatically create diagnosers and to integrate them into the system to identify if a failure has occurred. There are different methods of incorporating a diagnoser into a group of interacting services. One option is to modify the BPEL file representing services to incorporate the diagnoser. Another option is to implement the diagnoser as a separate service which interacts with the existing services. Moreover, the interaction between the diagnoser and the services can be either orchestration or choreography. As result, there are four options for the implementation of the diagnoser into the SoA via MDA. This paper reports on an Oracle JDeveloper plugin tool developed which applies MDA to create these four possible implementations and compares the performance of them with the help of a case study.
  • Keywords
    fault tolerant computing; software architecture; BPEL file representing services; Oracle JDeveloper plugin tool; choreography; diagnosers; fault tolerance; model driven architecture; orchestration; service oriented architectures; Automata; Computer architecture; Computer science; Condition monitoring; Discrete event systems; Fault tolerance; Fault tolerant systems; Service oriented architecture; Switches; Web services;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Enterprise Distributed Object Computing Conference Workshops, 2008 12th
  • Conference_Location
    Munich
  • Print_ISBN
    978-0-7695-3720-7
  • Type

    conf

  • DOI
    10.1109/EDOCW.2008.8
  • Filename
    4815030