Kepler: Robust Learning for Parametric Query Optimization
Author(s)
Doshi, Lyric
Zhuang, Vincent
Jain, Gaurav
Marcus, Ryan C
Huang, Haoyu
Alt?nb?ken, Deniz
Brevdo, Eugene
Fraser, Campbell
Date Issued
May 30, 2023
Journal
Proceedings of the ACM on Management of Data
Publisher
ACM
Citation
Doshi, Lyric, Zhuang, Vincent, Jain, Gaurav, Marcus, Ryan C, Huang, Haoyu et al. 2023. "Kepler: Robust Learning for Parametric Query Optimization." Proceedings of the ACM on Management of Data, 1 (1).
Version
Final published version
Abstract
Most existing parametric query optimization (PQO) techniques rely on traditional query optimizer cost models, which are often inaccurate and result in suboptimal query performance. We propose Kepler, an end-to-end learning-based approach to PQO that demonstrates significant speedups in query latency over a traditional query optimizer. Central to our method is Row Count Evolution (RCE), a novel plan generation algorithm based on perturbations in the sub-plan cardinality space. While previous approaches require accurate cost models, we bypass this requirement by evaluating candidate plans via actual execution data and training an ML model to predict the fastest plan given parameter binding values. Our models leverage recent advances in neural network uncertainty in order to robustly predict faster plans while avoiding regressions in query performance. Experimentally, we show that Kepler achieves significant improvements in query runtime on multiple datasets on PostgreSQL.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Creative Commons Attribution
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/3588963
