Recently enough data and analysis has become available to develop and evaluate diverse approaches for achieving and evaluating high reliability. It is now possible to connect them together in a wholistic way. This tutorial presents an integrated framework for software reliability techniques. There is a connection between the factors that influence defect density, test effectiveness and the reliability growth models. As developers move from phase to phase, they can find suitable methods for achieving and assessing reliability.

The focus is on the perspective and not on the details. While many results are taken from research publications, the tutorial is intended for practiceners. There is some mathematics in the tutorial, since it is engineering, but it is simple enough to be used with a spreadsheet. For those with heavy-duty needs, there will be a list of sophisticated tools available.

The topics include basic terminology, common reliability measures, test coverage measures, factors that control defect density, implications of software test methodologies, modeling software reliability growth, applying software reliability growth models, estimating residual defect density, coverage based modeling, reliability of multi-component Systems, tools for reliable software development.

Presenter: Yashwant Malaiya, Colorado State University

Yashwant K. Malaiya is a Professor of Computer Science at Colorado State University. He has published widely in the areas of fault modeling, software and hardware reliability, testing and testable design.

He has served as the General Chair of ISSRE '93 and as a Program Chair of ISSRE '97. He guest edited IEEE Software special issue on "Reliability Measurement", July 1992 and had edited "Software Reliability Models: Theoretical Development, Evaluation and Applications", (IEEE CS Press Technology Series), 1990. He has also been involved in several other IEEE CS conferences and society activities.

His current research interests include defect density estimation and control, test effectiveness and high reliability systems.