Learned Index Benefits: Machine Learning Based Index Performance Estimation

Summary: An end-to-end learned estimator predicts candidate-index benefits without optimizer “what-if” plans, using operation-aware features and attention to model index interactions. Transfer learning enables cross-database adaptation, improving estimation speed/accuracy and downstream index recommendations. (summarized by gpt-5.6-luna on Jul 24 2026)

Paper ID: h3318aca5db8907e7
Venue: VLDB
Year: 2022
Pagerank: 6.4329878e-05
Overall Rank: 4,726 | 68.33%
DOI: 10.14778/3565838.3565848
PDF: Download (CC BY-NC-ND 4.0)

Incoming Non-self Citations Over Time

Authors

1. Jiachen Shi (Institute for Infocomm Research; Nanyang Technological University)
2. Gao Cong (Nanyang Technological University)
3. Xiao-Li Li (Agency for Science, Technology and Research; Institute for Infocomm Research; Nanyang Technological University)

BibTeX Citation

@article{shi_vldb22,
        title = {{Learned Index Benefits: Machine Learning Based Index Performance Estimation}},
        author = {Shi, Jiachen and Cong, Gao and Li, Xiao-Li},
        journal = {PVLDB},
        series = {{VLDB} '22},
        volume = {15},
        number = {13},
        pages = {3950--3962},
        doi = {10.14778/3565838.3565848},
        url = {https://doi.org/10.14778/3565838.3565848},
        year = {2022}
}

Incoming Citations (Sorted by Pagerank)

Showing 15 of 15 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
6,121	Breaking It Down: An In-depth Study of Index Advisors	2024	VLDB	5.863893e-05
6,889	Accelerating String-key Learned Index Structures via Memoization-based Incremental Training	2024	VLDB	5.6393954e-05
7,238	Refactoring Index Tuning Process with Benefit Estimation	2024	VLDB	5.5624225e-05
7,997	The Holon Approach for Simultaneously Tuning Multiple Components in a Self-Driving Database Management System with Machine Learning via Synthesized Proto-Actions	2024	VLDB	5.3938336e-05
8,798	Automatic Indexing in Oracle	2025	VLDB	5.2606925e-05
9,829	Wii: Dynamic Budget Reallocation In Index Tuning	2024	SIGMOD	5.1066866e-05
10,184	Does A Fish Need a Bicycle? The Case for On-Chip NPUs in DBMS	2026	CIDR	5.052396e-05
10,323	This is Going to Sound Crazy, But What If We Used Large Language Models to Boost Automatic Database Tuning Algorithms By Leveraging Prior History? We Will Find Better Configurations More Quickly Than Retraining From Scratch!	2026	SIGMOD	5.0241309e-05
10,490	Are Learned DBMS Components Robust to Workload Drift?: [Experiments & Analysis]	2026	SIGMOD	4.9605343e-05
10,608	Rainbow: Risk-aware Index Benefit Estimation Facing Out Of Distribution Workloads	2026	SIGMOD	4.9605343e-05
10,676	Understanding and Detecting Query Performance Regression in Practical Index Tuning: [Experiments & Analysis]	2026	SIGMOD	4.9605343e-05
10,753	RIB: Robust Learning-based Index Benefit Estimation	2026	SIGMOD	4.9605343e-05
10,994	TATA: An Efficient Framework for Task Transfer in Query Plan Representation	2026	VLDB	4.9605343e-05
11,260	PLM4NDV: Minimizing Data Access for Number of Distinct Values Estimation with Pre-trained Language Models	2025	SIGMOD	4.9605343e-05
11,288	Esc: An Early-Stopping Checker for Budget-aware Index Tuning	2025	VLDB	4.9605343e-05

Outgoing Citations (Sorted by Pagerank)

Showing 19 of 19 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
15	How Good Are Query Optimizers, Really?	2016	VLDB	0.00061463175
84	Learned Cardinalities: Estimating Correlated Joins with Deep Learning	2019	CIDR	0.00035920657
142	Neo: A Learned Query Optimizer	2019	VLDB	0.00029114937
189	DB2 Design Advisor: Integrated Automatic Physical Database Design	2004	VLDB	0.00025846085
355	Bao: Making Learned Query Optimization Practical	2021	SIGMOD	0.00020146245
380	AutoAdmin "What-if" Index Analysis Utility	1998	SIGMOD	0.00019530546
453	An End-to-End Learning-based Cost Estimator	2020	VLDB	0.0001801482
473	The Making of TPC-DS	2006	VLDB	0.00017656775
545	Plan-Structured Deep Neural Network Models for Query Performance Prediction	2019	VLDB	0.0001657599
676	Robust Estimation of Resource Consumption for SQL Queries using Statistical Techniques	2012	VLDB	0.00014838681
1,261	Sampling-Based Query Re-Optimization	2016	SIGMOD	0.00011289724
1,282	AI Meets AI: Leveraging Query Executions to Improve Index Recommendations	2019	SIGMOD	0.00011197535
1,400	Magic mirror in my hand, which is the best in the land? An Experimental Evaluation of Index Selection Algorithms	2020	VLDB	0.00010759064
1,517	Automatically Indexing Millions of Databases in Microsoft Azure SQL Database	2019	SIGMOD	0.00010385462
1,823	CoPhy: A Scalable, Portable, and Interactive Index Advisor for Large Workloads	2011	VLDB	9.5457638e-05
2,036	Efficient Use of the Query Optimizer for Automated Physical Design	2007	VLDB	9.1356845e-05
2,888	Query Optimizers: Time to Rethink the Contract?	2009	SIGMOD	7.8895201e-05
3,852	Database Tuning Advisor for Microsoft SQL Server 2005: Demo	2005	SIGMOD	6.9614927e-05
4,629	Index Interactions in Physical Design Tuning: Modeling, Analysis, and Applications	2009	VLDB	6.4794914e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	834	Benchmarking Learned Indexes	2021	VLDB
2	9,365	Deep Query Optimization	2019	SIGMOD
3	10,988	Evaluating the Practical Effectiveness of LLM-Driven Index Tuning on Microsoft SQL Server	2026	VLDB
4	10,676	Understanding and Detecting Query Performance Regression in Practical Index Tuning: [Experiments & Analysis]	2026	SIGMOD
5	4,765	Machine Learning for Databases	2021	VLDB
6	1,282	AI Meets AI: Leveraging Query Executions to Improve Index Recommendations	2019	SIGMOD
7	7,238	Refactoring Index Tuning Process with Benefit Estimation	2024	VLDB
8	4,990	Budget-aware Index Tuning with Reinforcement Learning	2022	SIGMOD
9	10,753	RIB: Robust Learning-based Index Benefit Estimation	2026	SIGMOD
10	3,632	Learned Index: A Comprehensive Experimental Evaluation	2023	VLDB