An integrated approach for studying architectural evolution

Abstract

Studying how a software system has evolved over time is difficult, time consuming, and costly; existing techniques are often limited in their applicability, are hard to extend, and provide little support for coping with architectural change. The paper introduces an approach to studying software evolution that integrates the use of metrics, software visualization, and origin analysis, which is a set of techniques for reasoning about structural and architectural change. Our approach incorporates data from various statistical and metrics tools, and provides a query engine as well as a Web-based visualization and navigation interface. It aims to provide an extensible, integrated environment for aiding software maintainers in understanding the evolution of long-lived systems that have undergone significant architectural change. We use the evolution of GCC as an example to demonstrate the uses of various functionalities of BEAGLE, a prototype implementation of the proposed environment.