DobLIX: A Dual-Objective Learned Index for Log-Structured Merge Trees

Summary: DobLIX: a dual-objective learned index for LSM trees that trains models to minimize both index lookup error and downstream on-disk data-access cost, addressing I/O-heavy lookups when index parts live on disk. Includes an RL tuner for online adaptation, yielding 1.19×–2.21× throughput gains in RocksDB while preserving write efficiency. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID: 14203
Venue: VLDB
Year: 2025
Pagerank: 5.093636e-05
Overall Rank: 10,955 | 24.84%
DOI: 10.14778/3749646.3749667

Incoming Non-self Citations Over Time

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

Authors

1. Alireza Heidari (Huawei)
2. Amirhossein Ahmadi (Huawei)
3. Wei Zhang (Huawei)

BibTeX Citation

@article{heidari_vldb25,
        title = {{DobLIX: A Dual-Objective Learned Index for Log-Structured Merge Trees}},
        author = {Heidari, Alireza and Ahmadi, Amirhossein and Zhang, Wei},
        journal = {PVLDB},
        series = {{VLDB} '25},
        volume = {18},
        number = {11},
        pages = {3965--3978},
        doi = {10.14778/3749646.3749667},
        url = {https://doi.org/10.14778/3749646.3749667},
        year = {2025}
}

Incoming Citations (Sorted by Pagerank)

Showing 0 of 0 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank

Outgoing Citations (Sorted by Pagerank)

Showing 29 of 29 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
43	The Case for Learned Index Structures	2018	SIGMOD	0.00046060254
447	ALEX: An Updatable Adaptive Learned Index	2020	SIGMOD	0.00018322593
477	The PGM-index: a fully-dynamic compressed learned index with provable worst-case bounds	2020	VLDB	0.00017851226
790	FITing-Tree: A Data-aware Index Structure	2019	SIGMOD	0.0001401445
831	Dostoevsky: Better Space-Time Trade-Offs for LSM-Tree Based Key-Value Stores via Adaptive Removal of Superfluous Merging	2018	SIGMOD	0.00013748512
847	Benchmarking Learned Indexes	2021	VLDB	0.0001365768
885	Finding Frequent Items in Data Streams	2008	VLDB	0.00013419017
946	HoloDetect: Few-Shot Learning for Error Detection	2019	SIGMOD	0.00013054126
1,418	LISA: A Learned Index Structure for Spatial Data	2020	SIGMOD	0.00010835539
2,685	Approximate Denial Constraints	2020	VLDB	8.2594662e-05
2,806	Are Updatable Learned Indexes Ready?	2022	VLDB	8.1013097e-05
2,910	APEX: A High-Performance Learned Index on Persistent Memory	2022	VLDB	7.9700885e-05
3,729	CARMI: A Cache-Aware Learned Index with a Cost-based Construction Algorithm	2022	VLDB	7.1683974e-05
3,792	Learned Index: A Comprehensive Experimental Evaluation	2023	VLDB	7.1220982e-05
4,300	DILI: A Distribution-Driven Learned Index	2023	VLDB	6.773869e-05
4,578	PLIN: A Persistent Learned Index for Non-Volatile Memory with High Performance and Instant Recovery	2023	VLDB	6.6210545e-05
4,664	GRF: A Global Range Filter for LSM-Trees with Shape Encoding	2024	SIGMOD	6.5787544e-05
5,222	CliffGuard: A Principled Framework for Finding Robust Database Designs	2015	SIGMOD	6.3103741e-05
5,229	Dissecting, Designing, and Optimizing LSM-based Data Stores	2022	SIGMOD	6.307514e-05
5,423	Updatable Learned Indexes Meet Disk-Resident DBMS - From Evaluations to Design Choices	2023	SIGMOD	6.2242567e-05
5,766	Endure: A Robust Tuning Paradigm for LSM Trees Under Workload Uncertainty	2022	VLDB	6.094771e-05
6,687	Making In-Memory Learned Indexes Efficient on Disk	2024	SIGMOD	5.8022308e-05
6,843	Learning to Optimize LSM-trees: Towards A Reinforcement Learning based Key-Value Store for Dynamic Workloads	2023	SIGMOD	5.7573899e-05
7,224	LITS: An Optimized Learned Index for Strings	2024	VLDB	5.6675134e-05
7,693	CaaS-LSM: Compaction-as-a-Service for LSM-based Key-Value Stores in Storage Disaggregated Infrastructure	2024	SIGMOD	5.5663461e-05
8,156	How to Grow an LSM-tree? Towards Bridging the Gap Between Theory and Practice	2025	SIGMOD	5.4765648e-05
8,234	The Case for Learned In-Memory Joins	2023	VLDB	5.460955e-05
8,463	Limousine: Blending Learned and Classical Indexes to Self-Design Larger-than-Memory Cloud Storage Engines	2024	SIGMOD	5.4205593e-05
9,995	DumpKV: Learning based lifetime aware garbage collection for key value separation in LSM-tree	2025	VLDB	5.1814573e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	9,998	SWIX: A Memory-efficient Sliding Window Learned Index	2024	SIGMOD
2	4,300	DILI: A Distribution-Driven Learned Index	2023	VLDB
3	447	ALEX: An Updatable Adaptive Learned Index	2020	SIGMOD
4	6,843	Learning to Optimize LSM-trees: Towards A Reinforcement Learning based Key-Value Store for Dynamic Workloads	2023	SIGMOD
5	857	SlimDB: A Space-Efficient Key-Value Storage Engine For Semi-Sorted Data	2017	VLDB
6	7,459	Efficient Data Ingestion and Query Processing for LSM-Based Storage Systems	2019	VLDB
7	10,378	High Performance or Low Memory? An Updatable Learned Index Framework for Time-Space Tradeoff	2026	SIGMOD
8	10,617	LiBox: A Learned Index as an Array to Minimize Last-Mile Search	2026	VLDB
9	10,458	From Learning to Recycling: A Log-Structured Learned-Less Index	2026	SIGMOD
10	10,277	On Self-Designing Learned Indexes	2026	SIGMOD