A framework for classifying and comparing software architecture tools for quality evaluation

Abstract

Software quality is a crucial factor for system success. Several tools have been proposed to support the evaluation and comparison of software architecture designs. However, the diversity in focus, approaches, interfaces and results leaves the researcher and practitioner wondering what would be the most appropriate solution for their specific goals. This paper presents a comparison framework that identifies the most relevant features for categorizing different architecture evaluation tools according to six different dimensions. The results show the attributes that a comprehensive tool should support include: the ability to handle multiple modelling approaches, integration with the industry standard UML or specific ADL, support for trade-off analysis of competing quality attributes and, the reuse of knowledge through the build-up of new architectural patterns. This comparison is able to, not only guide the choice of evaluation, but also promote the development of more powerful tools for modeling and analysis of software architectures. © 2011 Springer-Verlag.