ArceKV: Towards Workload-driven LSM-compactions for Key-Value Store Under Dynamic Workloads

Summary: ArceKV targets dynamic KV workloads with workload-driven LSM compactions, moving beyond static read/write-ratio tuning. It relaxes traditional LSM constraints via ElasticLSM and uses the lightweight Arce decision engine to choose among compactions/write stalls, yielding ~3x gains on shifting workloads. (summarized by gpt-5.4-mini on Apr 12 2026)

Paper ID: 14623
Venue: VLDB
Year: 2026
Pagerank: 5.3428568e-05
Overall Rank: 8,843 | 39.54%
DOI: 10.14778/3796195.3796208

Incoming Non-self Citations Over Time

Authors

1. Junfeng Liu (Nanyang Technological University)
2. Haoxuan Xie (Nanyang Technological University)
3. Siqiang Luo (Nanyang Technological University)

BibTeX Citation

@article{liu_vldb26,
        title = {{ArceKV: Towards Workload-driven LSM-compactions for Key-Value Store Under Dynamic Workloads}},
        author = {Liu, Junfeng and Xie, Haoxuan and Luo, Siqiang},
        journal = {PVLDB},
        series = {{VLDB} '26},
        volume = {19},
        number = {5},
        pages = {958--972},
        doi = {10.14778/3796195.3796208},
        url = {https://doi.org/10.14778/3796195.3796208},
        year = {2026}
}

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
10,580	How to Write to SSDs	2026	VLDB	5.0723324e-05
10,701	Terark-DS: A High-Performance and Storage-Efficient Key-Value Separation Storage Engine on Disaggregated Storage	2026	VLDB	5.0723324e-05

Outgoing Citations (Sorted by Pagerank)

Showing 45 of 45 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
268	TiDB: A Raft-based HTAP Database	2020	VLDB	0.00022726088
273	bLSM: A General Purpose Log Structured Merge Tree	2012	SIGMOD	0.00022638589
438	Monkey: Optimal Navigable Key-Value Store	2017	SIGMOD	0.000184362
582	Workload-Aware Database Monitoring and Consolidation	2011	SIGMOD	0.00016109335
806	Dostoevsky: Better Space-Time Trade-Offs for LSM-Tree Based Key-Value Stores via Adaptive Removal of Superfluous Merging	2018	SIGMOD	0.00013854342
834	SlimDB: A Space-Efficient Key-Value Storage Engine For Semi-Sorted Data	2017	VLDB	0.0001370423
879	SuRF: Practical Range Query Filtering with Fast Succinct Tries	2018	SIGMOD	0.00013424048
1,302	X-Engine: An Optimized Storage Engine for Large-scale E-commerce Transaction Processing	2019	SIGMOD	0.00011217825
1,470	The Log-Structured Merge-Bush & the Wacky Continuum	2019	SIGMOD	0.00010664663
1,580	Compaction management in distributed key-value datastores	2015	VLDB	0.00010277211
1,670	Amazon DynamoDB: A Seamlessly Scalable Non-relational Datastore	2012	SIGMOD	0.00010029236
1,865	Design Continuums and the Path Toward Self-Designing Key-Value Stores that Know and Learn	2019	CIDR	9.5595832e-05
1,958	Chucky: A Succinct Cuckoo Filter for LSM-Tree	2021	SIGMOD	9.400672e-05
2,469	Viper: An Efficient Hybrid PMem-DRAM Key-Value Store	2021	VLDB	8.5204043e-05
2,696	Rosetta: A Robust Space-Time Optimized Range Filter for Key-Value Stores	2020	SIGMOD	8.2243064e-05
2,754	Lethe: A Tunable Delete-Aware LSM Engine	2020	SIGMOD	8.1411998e-05
2,918	Spooky: Granulating LSM-Tree Compactions Correctly	2022	VLDB	7.9389295e-05
2,921	OceanBase: A 707 Million tpmC Distributed Relational Database System	2022	VLDB	7.9374885e-05
2,933	Constructing and Analyzing the LSM Compaction Design Space	2021	VLDB	7.9207978e-05
2,963	Accordion: Better Memory Organization for LSM Key-Value Stores	2018	VLDB	7.8909953e-05
3,455	Cosine: A Cloud-Cost Optimized Self-Designing Key-Value Storage Engine	2022	VLDB	7.381231e-05
3,543	TreeLine: An Update-In-Place Key-Value Store for Modern Storage	2023	VLDB	7.3131121e-05
3,811	Leaper: A Learned Prefetcher for Cache Invalidation in LSM-tree based Storage Engines	2020	VLDB	7.0889353e-05
3,873	DINOMO: An Elastic, Scalable, High-Performance Key-Value Store for Disaggregated Persistent Memory	2022	VLDB	7.0412176e-05
4,128	Proteus: A Self-Designing Range Filter	2022	SIGMOD	6.8650026e-05
4,265	On Performance Stability in LSM-based Storage Systems	2020	VLDB	6.7775584e-05
4,680	GRF: A Global Range Filter for LSM-Trees with Shape Encoding	2024	SIGMOD	6.5512396e-05
4,857	Grafite: Taming Adversarial Queries with Optimal Range Filters	2024	SIGMOD	6.4569918e-05
4,920	Oasis: An Optimal Disjoint Segmented Learned Range Filter	2024	VLDB	6.4286711e-05
4,965	Lightweight Cardinality Estimation in LSM-based Systems	2018	SIGMOD	6.4076301e-05
5,031	A Comparative Study of Secondary Indexing Techniques in LSM-based NoSQL Databases	2018	SIGMOD	6.3771675e-05
5,236	Breaking Down Memory Walls: Adaptive Memory Management in LSM-based Storage Systems	2021	VLDB	6.2877974e-05
5,490	Endure: A Robust Tuning Paradigm for LSM Trees Under Workload Uncertainty	2022	VLDB	6.1855258e-05
5,512	Dotori: A Key-Value SSD Based KV Store	2023	VLDB	6.1786153e-05
6,414	Revisiting the Design of LSM-tree Based OLTP Storage Engine with Persistent Memory	2021	VLDB	5.8654384e-05
6,867	Learning to Optimize LSM-trees: Towards A Reinforcement Learning based Key-Value Store for Dynamic Workloads	2023	SIGMOD	5.7342475e-05
7,792	CAMAL: Optimizing LSM-trees via Active Learning	2024	SIGMOD	5.5276312e-05
8,194	How to Grow an LSM-tree? Towards Bridging the Gap Between Theory and Practice	2025	SIGMOD	5.4536597e-05
8,490	Limousine: Blending Learned and Classical Indexes to Self-Design Larger-than-Memory Cloud Storage Engines	2024	SIGMOD	5.3988256e-05
8,507	Mnemosyne: Dynamic Workload-Aware BF Tuning via Accurate Statistics in LSM trees	2025	SIGMOD	5.3944472e-05
8,927	Aster: Enhancing LSM-structures for Scalable Graph Database	2025	SIGMOD	5.3271923e-05
9,004	Rethinking The Compaction Policies in LSM-trees	2025	SIGMOD	5.3191308e-05
9,049	Structural Designs Meet Optimality: Exploring Optimized LSM-tree Structures in A Colossal Configuration Space	2024	SIGMOD	5.3091928e-05
9,576	Disco: A Compact Index for LSM-trees	2025	SIGMOD	5.231408e-05
10,115	DFlush: DPU-Offloaded Flush for Disaggregated LSM-based Key-Value Stores	2025	SIGMOD	5.1408306e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	2,933	Constructing and Analyzing the LSM Compaction Design Space	2021	VLDB
2	10,037	DumpKV: Learning based lifetime aware garbage collection for key value separation in LSM-tree	2025	VLDB
3	10,519	PartitionKV: Redesigning LSM-tree KV Stores on NVMs with Adaptive Partitioning for Reducing Write Stalls and Amplification	2026	SIGMOD
4	9,049	Structural Designs Meet Optimality: Exploring Optimized LSM-tree Structures in A Colossal Configuration Space	2024	SIGMOD
5	7,729	CaaS-LSM: Compaction-as-a-Service for LSM-based Key-Value Stores in Storage Disaggregated Infrastructure	2024	SIGMOD
6	834	SlimDB: A Space-Efficient Key-Value Storage Engine For Semi-Sorted Data	2017	VLDB
7	9,004	Rethinking The Compaction Policies in LSM-trees	2025	SIGMOD
8	10,510	Making LSM-Tree-based Key-Value Store Practical and Efficient for Multi-Tenant Serverless Cloud Databases	2026	SIGMOD
9	6,531	ArkDB: A Key-Value Engine for Scalable Cloud Storage Services	2021	SIGMOD
10	6,867	Learning to Optimize LSM-trees: Towards A Reinforcement Learning based Key-Value Store for Dynamic Workloads	2023	SIGMOD