Renovating Database Queries with Query AutoAwesome

Abstract

Querying is an important database activity, typically occurring more frequently than update. Programmers invest time and effort into developing a set of queries. We propose reusing these queries with Query AutoAwesome (QAA). QAA is a system to automatically enhance queries. QAA was inspired by Google's Auto Awesome tool, which provides automatic enhancements of photos. QAA ingests a query, the database schema, and the data to enhance a query. QAA can reuse a query in several ways, such as re-placing a literal with a choice of alternatives. From among the pool of potential enhancements it is important to choose the top-k generated queries. To do so we introduce objective functions to measure the enhancement and rank potential queries.