Paper Info
Title
Software Quality Analysis with Observation-Enhanced Quantitative Verification (Tutorial)
Abstract
Quantitative verification (QV) is a powerful tool for the analysis of performance, dependability and other quality properties of software systems. Supported by today's fast probabilistic model checkers, QV can analyse these properties for alternative designs and existing software, in domains ranging from service-based systems and cloud computing to embedded systems. Recent advances have dramatically improved the usefulness and accuracy of this analysis by exploiting observations of the software or its components, available for instance from logs, unit testing or monitoring. This tutorial will provide an introduction to the quantitative verification of software quality with the probabilistic model checker PRISM, followed by a presentation of two advanced techniques for observation-enhanced quantitative verification. The first technique computes confidence intervals for the analysed quality properties using parametric Markov models of the software system. The second technique refines the Markov models used to assess quality properties of component-based software by exploiting observations of the execution times of its components. Both techniques can significantly reduce the risk of invalid software engineering decisions, and are fully supported by new QV tools. The tutorial will include short exercises and practical demonstrations of PRISM and of the new QV tools. Attending it will benefit researchers and practitioners from the area of software performance and dependability engineering, as well as those interested in formal approaches to the modelling, analysis and verification of quality aspects of software.
Year
DOI
Venue
2017
10.1109/ICSAW.2017.37
2017 IEEE International Conference on Software Architecture Workshops (ICSAW)
Keywords
Field
DocType
software performance and reliability engineering,software quality,quantitative verification,stochastic models,Markov chain,probabilistic model checking
Software engineering,Computer science,Software system,Software metric,Software quality,Software verification and validation,Software construction,Reliability engineering,Software development,Software sizing,Software verification
Conference
ISBN
Citations 
PageRank 
978-1-5090-4794-9
0
0.34
References 
Authors
16
1
Name
Order
Citations
PageRank
Radu Calinescu190563.01