The effort and cost of maintenance dominate activities in a software system's lifecycle. Over the past two decades, software architecture research has yielded many tools and techniques for understanding and maintaining the architectures of software systems. However, empirical research and technology transfer are impeded by myriad disjoint environments, lack of a shared research infrastructure, the costs of making tools robust, and lack of datasets needed to conduct architecture-oriented empirical research. Consequently, researchers and practitioners often repeat each other's efforts, thus limiting potential breakthroughs. This projects develops the plans and an initial prototype for an integration framework that assembles architecture-related techniques to enable maintenance-focused empirical research. The framework revolutionizes the future of software systems research and development. It facilitates the discovery and adoption of cutting-edge techniques and tools, and fosters much more effective university-industry collaboration than exists today.
This planning project elicits and prioritizes the requirements, collects the community feedback, and develops a prototype for the Software Architecture INstrument (SAIN), a first-of-its-kind integration framework. SAIN comprises three principal components: (1) a catalogued library of tools for reverse engineering and analyzing software architectures; (2) a plug-and-play instrument for integrating the techniques to enable architecture-oriented empirical studies; and (3) a repository of Web-accessible software-architecture datasets and benchmarks. This planning project develops a proof-of-concept solution that demonstrates the feasibility of the proposed instrument, as well as the research activities enabled by this instrument. SAIN provides an effective, community-wide platform for software architecture research and practice through continuous integration of diverse tools. It supports architecture's central role in a broad range of software engineering activities. The framework comprises the first repository hosting large numbers of recovered architectures to enable researchers to effectively evaluate their techniques, and is aimed at reinvigorating empirical research in software architecture.
