Development of a Co-Evolution assistant to limit database decay

Abstract

There is currently a lack of informational tools to guide changes to database schemas in large-scale inormation systems. In practice, schema changes are made with relatively little regard for how they will impact the corresponding application code. Over time, this has been shown to cause an overall degradation of the quality of both the application code and the data stored in the database. We propose the Co-Evolution Assistant, a tool made specifically to fill this need for more impact visibility when making such structural changes in the database. Given a proposed schema change, we show how the Co-Evolution Assistant is able to provide a comprehensive analysis of the impact of that change, allowing system maintainers to make the best-informed decision. We then validate the performance of the tool on a real-world industrial codebase.