Using Genetic Search for Reverse Engineering of Parametric Behaviour Models for Performance Prediction

Created by W.Langdon from gp-bibliography.bib Revision:1.3949

@Article{Krogmann:2010:ieeeTSE,
  author =       "Klaus Krogmann and Michael Kuperberg and 
                 Ralf Reussner",
  title =        "Using Genetic Search for Reverse Engineering of
                 Parametric Behaviour Models for Performance
                 Prediction",
  journal =      "IEEE Transactions on Software Engineering",
  year =         "2010",
  month =        nov # "/" # dec,
  volume =       "36",
  number =       "6",
  pages =        "865--877",
  abstract =     "In component-based software engineering, existing
                 components are often reused in new applications.
                 Correspondingly, the response time of an entire
                 component-based application can be predicted from the
                 execution durations of individual component services.
                 These execution durations depend on the run time
                 behaviour of a component which itself is influenced by
                 three factors: the execution platform, the usage
                 profile, and the component wiring. To cover all
                 relevant combinations of these influencing factors,
                 conventional prediction of response times requires
                 repeated deployment and measurements of component
                 services for all such combinations, incurring a
                 substantial effort. This paper presents a novel
                 comprehensive approach for reverse engineering and
                 performance prediction of components. In it, genetic
                 programming is used for reconstructing a behavior model
                 from monitoring data, runtime bytecode counts, and
                 static bytecode analysis. The resulting behavior model
                 is parametrised over all three performance-influencing
                 factors, which are specified separately. This results
                 in significantly fewer measurements: The behaviour
                 model is reconstructed only once per component service,
                 and one application-independent bytecode benchmark run
                 is sufficient to characterise an execution platform. To
                 predict the execution durations for a concrete
                 platform, our approach combines the behaviour model
                 with platform-specific benchmarking results. We
                 validate our approach by predicting the performance of
                 a file sharing application.",
  keywords =     "genetic algorithms, genetic programming, sbse,
                 application independent bytecode benchmark, component
                 based software engineering, genetic search, parametric
                 behaviour model, reverse engineering, runtime bytecode
                 count, static bytecode analysis, object-oriented
                 programming, reverse engineering, search problems,
                 software performance evaluation",
  DOI =          "doi:10.1109/TSE.2010.69",
  ISSN =         "0098-5589",
  notes =        "p868 'overhead of BYCOUNTER ... at most 250 percent'
                 p869 SVM 'do not lend themselves easily to human
                 understanding'. JGAP p870 restricted ranges for ERC
                 known as 'special constants' see
                 \cite{daida:2001:GPEM}. p873 'Through static code
                 analysis, a gene representing the constant XXX extract
                 from bytecode was available to genetic programming'.
                 MARS 1500 generations 'applies genetic programming for
                 each byte code instruction' palladiofileshare p874 in
                 98percent GP approximations better than MARS
                 approximations. Also known as \cite{5530323}",
}

Genetic Programming entries for Klaus Krogmann Michael Kuperberg Ralf H Reussner

Citations