Read more
Zusatztext ? written by a scientist successful in performance evaluation! it is based on his experience and provides many ideas not only to laymen entering the field! but also to practitioners looking for inspiration. The work can be read systematically as a textbook on how to model and test the derived hypotheses on the basis of simulations. Also! separate parts can be studied! as the chapters are self-contained. ? the book can be successfully used either for self-study or as a supplementary book for a lecture. I believe that different types of readers will like it: practicing engineers and researchers dealing with new solutions! as well as graduate students starting their adventures in the jungle of performance evaluation. -Piotr Cholda! in IEEE Communications Magazine! October 2011 Informationen zum Autor Jean-Yves Le Boudec is professor at EPFL and fellow of the IEEE. He graduated from Ecole Normale Superieure de Saint-Cloud! Paris! and received his doctorate in 1984 from the University of Rennes! France. Before joining the EPFL in 1994! he had worked at Bell Northern Research! Ottawa! Canada; and at the IBM Zurich Research Laboratory where he was manager of the Customer Premises Network Department. His interests are in the performance and architecture of communication systems. He is recipient of numerous awards! including the 2008 IEEE Communications Society William R. Bennett Prize in the field of Communications Networking. This book is written for computer engineers and scientists active in the development of software and hardware systems. It supplies the understanding and tools needed to effectively evaluate the performance of individual computer and communication systems. It covers the theoretical foundations of the field as well as specific software packages being employed by leaders in the field. written by a scientist successful in performance evaluation, it is based on his experience and provides many ideas not only to laymen entering the field, but also to practitioners looking for inspiration. The work can be read systematically as a textbook on how to model and test the derived hypotheses on the basis of simulations. Also, separate parts can be studied, as the chapters are self-contained. the book can be successfully used either for self-study or as a supplementary book for a lecture. I believe that different types of readers will like it: practicing engineers and researchers dealing with new solutions, as well as graduate students starting their adventures in the jungle of performance evaluation. Piotr Cholda, in IEEE Communications Magazine, October 2011 Zusammenfassung This book is written for computer engineersand scientists active in the development of software andhardware systems. Itsupplies theunderstanding andtools needed to effectively evaluate the performance of individual computer and communication systems. Itcoversthe theoretical foundations of the fieldas well asspecific software packages being employed by leaders in the field. Inhaltsverzeichnis MethodologyWhat is Performance Evaluation ? Factors Evaluation Methods The Scientific Method Performance PatternsSummarizing Performance Data! Confidence IntervalsSummarized Performance DataConfidence IntervalsThe Independence AssumptionPrediction IntervalWhich Summarization To Use?Other Aspects of Confidence/Prediction IntervalsProofsModel FittingModel Fitting CriteriaLinear Regression Linear Regression withNorm Minimization Choosing a DistributionHeavy TailProofsTestsThe Neyman Pearson FrameworkLikelihood Ratio TestsANOVAAsymptotic ResultsOther TestsProofsForecastingWhat is Forecasting ? Linear Regression The Overfitting ProblemDifferencing the DataFitting Differenced Data to an ARMA ModelSparse ARMA and ARIMA ModelsProofsDiscrete Event SimulationWhat is a Simulation?Simulation TehniquesComputing the Accuracy of Stochastic SimulationsMonte Carlo SimulationR...
