ReoptSMART : A Learning Query Plan Cache Columbia University Computer Science Technical Report cucs-023-08

The task of query optimization in modern relational database systems is important but can be computationally expensive. Parametric query optimization(PQO) has as its goal the prediction of optimal query execution plans based on historical results, without consulting the query optimizer. We develop machine learning techniques that can accurately model the output of a query optimizer. Our algorithms handle non-linear boundaries in plan space and achieve high prediction accuracy even when a limited amount of data is available for training. We use both predicted and actual query execution times for learning, and are the first to demonstrate a total net win of a PQO method over a state-of-the-art query optimizer for some workloads. ReoptSMART realizes savings not only in optimization time, but also in query execution time, for an over-all improvement by more than an order of magnitude in some cases.