PAR2QO: Parametric Penalty-Aware Robust Query Optimization

Parametric Query Optimization (PQO) is an important problem in database systems, yet existing approaches suffer from high training costs, sensitivity to estimation errors, and vulnerability to severe performance regressions. This paper introduces PAR2QO (PARametric Penalty-Aware Robust Query Optimization), a system that integrates robust query optimization into PQO. PAR2QO strategically obtains plans from a well-balanced set of probe locations informed by the workload, and caches them as plan-penalty profiles. At runtime, PAR2QO selects the plan with the lowest expected penalty, explicitly accounting for selectivity uncertainties. Extensive experiments show that PAR2QO delivers significant speedups over existing methods while ensuring robustness against performance degradation. Additionally, we introduce CARVER, a workload generator aimed at covering possible cardinalities of subqueries. Not only does CARVER provide a more comprehensive way to evaluate PQO methods, but when used for training learned methods, it can also enhance their generalizability and stability.

Duke Scholars

Author Haibo Xiu  

Author Jun Yang Computer Science

Author Pankaj K. Agarwal Computer Science