Metrics -guided models and methods for cost and quality management of component -based software
Abstract
The paradigm shift to commercial off-the-shelf software components appears inevitable, necessitating drastic changes to current software development and evolution practices. Lack of confidence in the quality of third-party software components and the systems based on them has currently limited the applicability of this efficient paradigm to non-critical applications. Software metrics can be used to objectively quantify the quality of software components and systems, alleviating quality and risk concerns and raising assurance in component-based systems. This research identifies a set of software metrics pertinent to cost and quality management of component-based systems. To determine the extent of testing required for quality assurance, a temporal model is developed for predicting the value of test coverage, which is one of the proposed metrics. The metrics are then utilized in developing a graphical model for characterization of a component-based software system based on quality attributes of its constituent components and integration code. Based on this model, a development methodology is proposed for guiding acquisition and integration efforts by using multi-objective optimization to select components that will yield the highest quality within affordable cost. Enterprise integration, which aims at aligning business operations and information technology resources in an enterprise, is an emerging application of the component-based paradigm, and well-suited to metrics-guided optimization. As a final contribution of the research, the aforementioned techniques are applied to cost and quality management of enterprise integration.
Degree
Ph.D.
Advisors
Ghafoor, Purdue University.
Subject Area
Electrical engineering|Computer science|Systems design
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.