Approximate Query Processing with Error Guarantees

In recent years, with the increase of data and the sophistication of analysis requirements, query processing in databases has become more important. Recently, approximate query processing (AQP) was proposed for efficiently executing database queries on big data. In this research, we focus on synopsis construction on a relational database and the query technology based on it, which is called Bounded Approximate Query (BAQ) proposed in 2019. BAQ is a synopsis construction method that focuses on aggregate queries using SQL, and realizes error-guaranteed query processing by grouping the dataset into the synopsis. In this paper, we point out the limitations of queries and datasets in BAQ and based on the result of experiments, we prove that the proposed method can be applied efficiently to data wider than the original BAQ with smaller synopsis within the error guarantee.