Eraser: Eliminating Performance Regression on Learned Query Optimizer
Summary: Eraser removes performance regressions in learned query optimizers by estimating per-plan prediction reliability with a two-stage approach: a coarse filter for unseen features and cluster-based fine-grained reliability scoring. Pluggable across systems (Postgres, Spark), it preserves learned-optimizer gains while largely eliminating regressions and adapting to dynamic workloads. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Lianggui Weng (Alibaba)
- 2. Rong Zhu (Alibaba)
- 3. Di Wu (Alibaba; Huazhong University of Science and Technology)
- 4. Bolin Ding (Alibaba)
- 5. Bolong Zheng (Huazhong University of Science and Technology)
- 6. Jingren Zhou (Alibaba)
BibTeX Citation
@article{weng_vldb24,
title = {{Eraser: Eliminating Performance Regression on Learned Query Optimizer}},
author = {Weng, Lianggui and Zhu, Rong and Wu, Di and Ding, Bolin and Zheng, Bolong and Zhou, Jingren},
journal = {PVLDB},
series = {{VLDB} '24},
volume = {17},
number = {5},
pages = {926--938},
doi = {10.14778/3641204.3641205},
url = {https://doi.org/10.14778/3641204.3641205},
year = {2024}
}Incoming Citations (Sorted by Pagerank)
Showing 11 of 11 citing papers.
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 26 of 26 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 4,202 | Kepler: Robust Learning for Faster Parametric Query Optimization | 2023 | SIGMOD |
| 2 | 142 | Neo: A Learned Query Optimizer | 2019 | VLDB |
| 3 | 2,696 | Cost-based or Learning-based? A Hybrid Query Optimizer for Query Plan Selection | 2022 | VLDB |
| 4 | 3,212 | Efficiently Approximating Selectivity Functions using Low Overhead Regression Models | 2020 | VLDB |
| 5 | 6,584 | Can Large Language Models Be Query Optimizer for Relational Databases? | 2026 | SIGMOD |
| 6 | 10,764 | Practical Parameterized Query Optimization via Efficient Plan Reuse and List-wise Ranking | 2026 | SIGMOD |
| 7 | 6,109 | Is Your Learned Query Optimizer Behaving As You Expect? A Machine Learning Perspective | 2024 | VLDB |
| 8 | 2,190 | Lero: A Learning-to-Rank Query Optimizer | 2023 | VLDB |
| 9 | 6,672 | Speeding Up End-to-end Query Execution via Learning-based Progressive Cardinality Estimation | 2023 | SIGMOD |
| 10 | 3,481 | LOGER: A Learned Optimizer towards Generating Efficient and Robust Query Execution Plans | 2023 | VLDB |