A Review of Cardinality Estimation Using Machine Learning Methods

Abstract

Query optimizer (QO) is one of the most significant components in relational database
management system (DBMS) that finds the best route to execute the query. Thus, it has a
significant role in the efficiency of DBMS. It is necessary to correctly estimate the
optimization of QO cardinality under the queries and the conditional clauses in them to
increase accuracy and efficiency. In recent years, much attention has been paid to using of
machine learning methods to estimate cardinality. The paper reviewed the activities
concerning using machine learning to solve the problem of cardinality estimation and
optimization of query. As most of the studies done in this area have used neural networks
and reinforcement learning (RL), the present paper has focused on these works and while
examining and comparing them, has also analyzed the challenges ahead