Learning to Optimize LSM-trees: Towards A Reinforcement Learning based Key-Value Store for Dynamic Workloads

Summary: RusKey pioneers online LSM-tree design and RL-guided LSM transformations for dynamic workloads. FLSM-tree enables efficient transitions between compaction policies without workload priors, delivering up to 4× end-to-end performance vs RocksDB. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: hfbf619f14249df0d
Venue: SIGMOD
Year: 2023
Pagerank: 5.8771312e-05
Overall Rank: 6,123 | 58.85%
DOI: 10.1145/3617333

Incoming Non-self Citations Over Time

Authors

1. Dingheng Mo (Nanyang Technological University)
2. Fanchao Chen (Fudan University)
3. Siqiang Luo (Nanyang Technological University)
4. Caihua Shan (Microsoft)

BibTeX Citation

@inproceedings{mo_sigmod23,
        title = {{Learning to Optimize LSM-trees: Towards A Reinforcement Learning based Key-Value Store for Dynamic Workloads}},
        author = {Mo, Dingheng and Chen, Fanchao and Luo, Siqiang and Shan, Caihua},
        series = {{SIGMOD} '23},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3617333},
        url = {https://dl.acm.org/doi/10.1145/3617333},
        year = {2023}
}

Incoming Citations (Sorted by Pagerank)

Showing 16 of 16 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
4,753	GRF: A Global Range Filter for LSM-Trees with Shape Encoding	2024	SIGMOD	6.4334982e-05
4,878	Oasis: An Optimal Disjoint Segmented Learned Range Filter	2024	VLDB	6.369385e-05
7,909	CAMAL: Optimizing LSM-trees via Active Learning	2024	SIGMOD	5.4266123e-05
8,228	ArceKV: Towards Workload-driven LSM-compactions for Key-Value Store Under Dynamic Workloads	2026	VLDB	5.3734043e-05
8,761	A New Paradigm in Tuning Learned Indexes: A Reinforcement Learning Enhanced Approach	2025	SIGMOD	5.2821881e-05
9,064	Aster: Enhancing LSM-structures for Scalable Graph Database	2025	SIGMOD	5.2270607e-05
9,140	Rethinking The Compaction Policies in LSM-trees	2025	SIGMOD	5.2208299e-05
9,191	Structural Designs Meet Optimality: Exploring Optimized LSM-tree Structures in A Colossal Configuration Space	2024	SIGMOD	5.20942e-05
9,654	Are Joins over LSM-trees Ready? Take RocksDB as an Example	2025	VLDB	5.142891e-05
10,307	DobLIX: A Dual-Objective Learned Index for Log-Structured Merge Trees	2025	VLDB	5.0392037e-05
10,491	Mathematical Foundations of Poisoning Attacks on Linear Regression over Cumulative Distribution Functions	2026	SIGMOD	4.9769913e-05
10,663	Improving Range Scan Performance in LSM-trees with Group Caching	2026	SIGMOD	4.9769913e-05
10,854	Rethinking Learned Index and LSM-tree Integration	2026	VLDB	4.9769913e-05
10,865	How Much Can RocksDB Chew? Achieving Near-Zero Write Stalls with Sustainable RocksDB	2026	VLDB	4.9769913e-05
11,436	AXE: A Task Decomposition Approach to Learned LSM Tuning	2025	VLDB	4.9769913e-05
11,582	TIGER: Training Inductive Graph Neural Network for Large-scale Knowledge Graph Reasoning	2024	VLDB	4.9769913e-05

Outgoing Citations (Sorted by Pagerank)

Showing 33 of 33 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
40	The Case for Learned Index Structures	2018	SIGMOD	0.00046363107
78	Automatic Database Management System Tuning Through Large-scale Machine Learning	2017	SIGMOD	0.00036675568
224	Self-Driving Database Management Systems	2017	CIDR	0.00024011047
236	LinkBench: a Database Benchmark Based on the Facebook Social Graph	2013	SIGMOD	0.00023664907
258	bLSM: A General Purpose Log Structured Merge Tree	2012	SIGMOD	0.00022932099
314	An End-to-End Automatic Cloud Database Tuning System Using Deep Reinforcement Learning	2019	SIGMOD	0.00021276452
400	Monkey: Optimal Navigable Key-Value Store	2017	SIGMOD	0.00019124757
437	QTune: A Query-Aware Database Tuning System with Deep Reinforcement Learning	2019	VLDB	0.00018310278
461	Query-based Workload Forecasting for Self-Driving Database Management Systems	2018	SIGMOD	0.00017841988
585	Workload-Aware Database Monitoring and Consolidation	2011	SIGMOD	0.00015942562
753	Dostoevsky: Better Space-Time Trade-Offs for LSM-Tree Based Key-Value Stores via Adaptive Removal of Superfluous Merging	2018	SIGMOD	0.00014232329
817	SlimDB: A Space-Efficient Key-Value Storage Engine For Semi-Sorted Data	2017	VLDB	0.00013685662
889	SuRF: Practical Range Query Filtering with Fast Succinct Tries	2018	SIGMOD	0.00013243846
921	AsterixDB: A Scalable, Open Source BDMS	2014	VLDB	0.00013064043
1,289	An Inquiry into Machine Learning-based Automatic Configuration Tuning Services on Real-World Database Management Systems	2021	VLDB	0.00011159167
1,313	X-Engine: An Optimized Storage Engine for Large-scale E-commerce Transaction Processing	2019	SIGMOD	0.00011055196
1,422	The Log-Structured Merge-Bush & the Wacky Continuum	2019	SIGMOD	0.00010720711
1,569	Compaction management in distributed key-value datastores	2015	VLDB	0.00010207173
1,891	Design Continuums and the Path Toward Self-Designing Key-Value Stores that Know and Learn	2019	CIDR	9.4233024e-05
1,982	Chucky: A Succinct Cuckoo Filter for LSM-Tree	2021	SIGMOD	9.2570284e-05
2,664	Rosetta: A Robust Space-Time Optimized Range Filter for Key-Value Stores	2020	SIGMOD	8.1532061e-05
2,771	Constructing and Analyzing the LSM Compaction Design Space	2021	VLDB	8.0338932e-05
2,788	Lethe: A Tunable Delete-Aware LSM Engine	2020	SIGMOD	8.0114055e-05
2,868	Spooky: Granulating LSM-Tree Compactions Correctly	2022	VLDB	7.9205462e-05
2,996	Accordion: Better Memory Organization for LSM Key-Value Stores	2018	VLDB	7.7670057e-05
3,233	TreeLine: An Update-In-Place Key-Value Store for Modern Storage	2023	VLDB	7.5008192e-05
3,480	Cosine: A Cloud-Cost Optimized Self-Designing Key-Value Storage Engine	2022	VLDB	7.2661848e-05
3,672	Leaper: A Learned Prefetcher for Cache Invalidation in LSM-tree based Storage Engines	2020	VLDB	7.1084283e-05
4,198	Proteus: A Self-Designing Range Filter	2022	SIGMOD	6.7385809e-05
4,943	A Comparative Study of Secondary Indexing Techniques in LSM-based NoSQL Databases	2018	SIGMOD	6.3423449e-05
5,022	Breaking Down Memory Walls: Adaptive Memory Management in LSM-based Storage Systems	2021	VLDB	6.309271e-05
5,047	Lightweight Cardinality Estimation in LSM-based Systems	2018	SIGMOD	6.2972183e-05
5,250	Endure: A Robust Tuning Paradigm for LSM Trees Under Workload Uncertainty	2022	VLDB	6.2092552e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	10,669	Making LSM-Tree-based Key-Value Store Practical and Efficient for Multi-Tenant Serverless Cloud Databases	2026	SIGMOD
2	12,214	Splaying Log-Structured Merge-Trees	2018	SIGMOD
3	10,182	DumpKV: Learning based lifetime aware garbage collection for key value separation in LSM-tree	2025	VLDB
4	817	SlimDB: A Space-Efficient Key-Value Storage Engine For Semi-Sorted Data	2017	VLDB
5	6,639	LSM-Trees and B-Trees: The Best of Both Worlds	2019	SIGMOD
6	9,140	Rethinking The Compaction Policies in LSM-trees	2025	SIGMOD
7	753	Dostoevsky: Better Space-Time Trade-Offs for LSM-Tree Based Key-Value Stores via Adaptive Removal of Superfluous Merging	2018	SIGMOD
8	7,909	CAMAL: Optimizing LSM-trees via Active Learning	2024	SIGMOD
9	9,191	Structural Designs Meet Optimality: Exploring Optimized LSM-tree Structures in A Colossal Configuration Space	2024	SIGMOD
10	8,228	ArceKV: Towards Workload-driven LSM-compactions for Key-Value Store Under Dynamic Workloads	2026	VLDB