Tidehunter: Large-Value Storage With Minimal Data Relocation

Summary: Tidehunter makes the WAL permanent, avoiding value compaction for large-value LSM workloads via lazy key-to-position indexes and background relocation. Lock-free NVMe writes deliver 6.8× RocksDB throughput and stable production performance in Sui. (summarized by gpt-5.6-luna on Aug 28 2026)

Paper ID: hb2252916a49bd522
Venue: VLDB
Year: 2026
Pagerank: 4.9793485e-05
Overall Rank: 10,897 | 26.74%
DOI: 10.14778/3836663.3836725

Incoming Non-self Citations Over Time

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

Authors

1. Andrey Chursin (Mysten Labs)
2. Lefteris Kokoris-Kogias (Mysten Labs)
3. Alex Orlov (Mysten Labs)
4. Alberto Sonnino (Mysten Labs; University College London)
5. Igor Zablotchi (Mysten Labs)

BibTeX Citation

@article{chursin_vldb26,
        title = {{Tidehunter: Large-Value Storage With Minimal Data Relocation}},
        author = {Chursin, Andrey and Kokoris-Kogias, Lefteris and Orlov, Alex and Sonnino, Alberto and Zablotchi, Igor},
        journal = {PVLDB},
        series = {{VLDB} '26},
        volume = {19},
        number = {11},
        pages = {3786--3798},
        doi = {10.14778/3836663.3836725},
        url = {https://doi.org/10.14778/3836663.3836725},
        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 11 of 11 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
35	Hekaton: SQL Server’s Memory-Optimized OLTP Engine	2013	SIGMOD	0.00048001919
40	The Case for Learned Index Structures	2018	SIGMOD	0.00046284649
63	Amazon Aurora: Design Considerations for High Throughput Cloud-Native Relational Databases	2017	SIGMOD	0.00038531147
400	Monkey: Optimal Navigable Key-Value Store	2017	SIGMOD	0.00019129175
636	ERMIA: Fast Memory-Optimized Database System for Heterogeneous Workloads	2016	SIGMOD	0.00015360975
754	Dostoevsky: Better Space-Time Trade-Offs for LSM-Tree Based Key-Value Stores via Adaptive Removal of Superfluous Merging	2018	SIGMOD	0.00014236015
1,399	LLAMA: A Cache/Storage Subsystem for Modern Hardware	2013	VLDB	0.00010781976
1,437	Write-Behind Logging	2017	VLDB	0.0001065408
3,232	TreeLine: An Update-In-Place Key-Value Store for Modern Storage	2023	VLDB	7.50343e-05
5,506	Managing ML Pipelines: Feature Stores and the Coming Wave of Embedding Ecosystems	2021	VLDB	6.1016716e-05
7,544	The Hopsworks Feature Store for Machine Learning	2024	SIGMOD	5.4983875e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	2,788	Lethe: A Tunable Delete-Aware LSM Engine	2020	SIGMOD
2	11,863	Workload-Adaptive Filtering in Storage Engines	2022	SIGMOD
3	3,680	SplinterDB and Maplets: Improving the Tradeoffs in Key-Value Store Compaction Policy	2023	SIGMOD
4	8,221	ArceKV: Towards Workload-driven LSM-compactions for Key-Value Store Under Dynamic Workloads	2026	VLDB
5	9,035	ChainKV: A Semantics-Aware Key-Value Store for Ethereum System	2023	SIGMOD
6	10,856	How Much Can RocksDB Chew? Achieving Near-Zero Write Stalls with Sustainable RocksDB	2026	VLDB
7	10,933	SOS: A High-Performance Distributed Key-Value Store for Large-Scale Online Services	2026	VLDB
8	3,232	TreeLine: An Update-In-Place Key-Value Store for Modern Storage	2023	VLDB
9	9,690	FluidKV: Seamlessly Bridging the Gap between Indexing Performance and Memory-Footprint on Ultra-Fast Storage	2024	VLDB
10	816	SlimDB: A Space-Efficient Key-Value Storage Engine For Semi-Sorted Data	2017	VLDB