• DocumentCode
    1931966
  • Title

    PerfExt: Performance Extrapolation Tool

  • Author

    Duttagupta, Subhasri ; Virk, Rupinder Singh

  • Author_Institution
    TCS Innovation Labs., Tata Consultancy Services, Mumbai, India
  • fYear
    2012
  • fDate
    25-27 Sept. 2012
  • Firstpage
    369
  • Lastpage
    374
  • Abstract
    Load testing of IT applications is fraught with the challenges of time to market, quality of results, high cost of commercial tools, and accurately representing production like scenarios. It would help IT projects to be able to test with a small number of users and extrapolate to scenarios with much larger number of users. This in turn will cut down cycle times and costs and allow for a variety of extrapolations closer to production. We present Perf Ext, a performance extrapolation tool that uses a statistical empirical modeling technique to extrapolate application performance for a large number of users. The tool provides more than 90% accuracy for a range of applications running across a number of hardware servers. The tool allows performance extrapolation of applications with multiple transactions and its performance is compared with that of MVA.
  • Keywords
    extrapolation; program testing; statistical analysis; PerfExt; load testing; performance extrapolation tool; statistical empirical modeling technique; Analytical models; Extrapolation; Load modeling; Servers; Testing; Throughput; Time factors; extrapolation; load testing; performance; regression;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computational Intelligence, Modelling and Simulation (CIMSiM), 2012 Fourth International Conference on
  • Conference_Location
    Kuantan
  • ISSN
    2166-8531
  • Print_ISBN
    978-1-4673-3113-5
  • Type

    conf

  • DOI
    10.1109/CIMSim.2012.71
  • Filename
    6338106