Libra: One-Shot Parameter Sensitivity Estimation for Transfer Learning in Database Performance Prediction

Summary: Libra is a transfer-learning framework for DBMS performance prediction that predicts a target context’s parameter-sensitivity profile in one shot, then retrieves the most similar source context. It avoids negative transfer by focusing sampling on high-impact parameters, yielding up to 32x less sampling and large error reductions across 161 contexts. (summarized by gpt-5.4-mini on Apr 12 2026)

Paper ID: 14572
Venue: VLDB
Year: 2026
Pagerank: 5.093636e-05
Overall Rank: 10,626 | 27.10%
DOI: 10.14778/3796195.3796207

Incoming Non-self Citations Over Time

No non-self incoming citations found for this paper in this database.

Authors

1. Tatsuhiro Nakamori (Keio University)
2. Hideyuki Kawashima (Keio University)

BibTeX Citation

@article{nakamori_vldb26,
        title = {{Libra: One-Shot Parameter Sensitivity Estimation for Transfer Learning in Database Performance Prediction}},
        author = {Nakamori, Tatsuhiro and Kawashima, Hideyuki},
        journal = {PVLDB},
        series = {{VLDB} '26},
        volume = {19},
        number = {5},
        pages = {945--957},
        doi = {10.14778/3796195.3796207},
        url = {https://doi.org/10.14778/3796195.3796207},
        year = {2026}
}

Incoming Citations (Sorted by Pagerank)

Showing 0 of 0 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank

Outgoing Citations (Sorted by Pagerank)

Showing 15 of 15 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
86	Automatic Database Management System Tuning Through Large-scale Machine Learning	2017	SIGMOD	0.00035316107
334	An End-to-End Automatic Cloud Database Tuning System Using Deep Reinforcement Learning	2019	SIGMOD	0.00020875082
347	Tuning Database Configuration Parameters with iTuned	2009	VLDB	0.00020651582
563	Plan-Structured Deep Neural Network Models for Query Performance Prediction	2019	VLDB	0.0001650812
1,344	An Inquiry into Machine Learning-based Automatic Configuration Tuning Services on Real-World Database Management Systems	2021	VLDB	0.00011094717
2,355	QueryFormer: A Tree Transformer Model for Query Plan Representation	2022	VLDB	8.7022189e-05
2,740	Facilitating Database Tuning with Hyper-Parameter Optimization: A Comprehensive Experimental Evaluation	2022	VLDB	8.1855759e-05
2,844	Zero-Shot Cost Models for Out-of-the-box Learned Cost Prediction	2022	VLDB	8.0608767e-05
3,116	ResTune: Resource Oriented Tuning Boosted by Meta-Learning for Cloud Databases	2021	SIGMOD	7.7390737e-05
3,343	LlamaTune: Sample-Efficient DBMS Configuration Tuning	2022	VLDB	7.4983591e-05
4,011	Towards Dynamic and Safe Configuration Tuning for Cloud Databases	2022	SIGMOD	6.959982e-05
5,107	Stage: Query Execution Time Prediction in Amazon Redshift	2024	SIGMOD	6.3623786e-05
5,171	An Efficient Transfer Learning Based Configuration Adviser for Database Tuning	2024	VLDB	6.3347618e-05
8,572	T3: Accurate and Fast Performance Prediction for Relational Database Systems With Compiled Decision Trees	2025	SIGMOD	5.4102362e-05
9,016	DBSeer: Pain-free Database Administration through Workload Intelligence	2015	VLDB	5.3312701e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	3,343	LlamaTune: Sample-Efficient DBMS Configuration Tuning	2022	VLDB
2	9,441	Db2une: Tuning Under Pressure via Deep Learning	2024	VLDB
3	2,844	Zero-Shot Cost Models for Out-of-the-box Learned Cost Prediction	2022	VLDB
4	6,323	Modeling Shifting Workloads for Learned Database Systems	2024	SIGMOD
5	563	Plan-Structured Deep Neural Network Models for Query Performance Prediction	2019	VLDB
6	4,643	Learned Index Benefits: Machine Learning Based Index Performance Estimation	2022	VLDB
7	5,171	An Efficient Transfer Learning Based Configuration Adviser for Database Tuning	2024	VLDB
8	5,073	Database Workload Characterization with Query Plan Encoders	2022	VLDB
9	4,468	One Model to Rule them All: Towards Zero-Shot Learning for Databases	2022	CIDR
10	6,357	A Unified Transferable Model for ML-Enhanced DBMS	2022	CIDR