DBScholar

Back to papers

How Much Can RocksDB Chew? Achieving Near-Zero Write Stalls with Sustainable RocksDB

Summary: S-RocksDB reframes LSM write stalls as a continuous-control problem, using online reinforcement learning to match foreground admission with time-varying compaction capacity. Safe-state guardrails nearly eliminate stalls (69 vs. 64.3M) while bounding P99.99 latency below 0.11 ms. (summarized by gpt-5.6-luna on Aug 28 2026)

Paper ID
hbb7de41abbcfe0eb
Venue
VLDB
Year
2026
Pagerank
4.9793485e-05
Overall Rank
10,856 | 27.02%
DOI
10.14778/3836663.3836668

Incoming Non-self Citations Over Time

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

Authors

BibTeX Citation

@article{shin_vldb26,
        title = {{How Much Can RocksDB Chew? Achieving Near-Zero Write Stalls with Sustainable RocksDB}},
        author = {Shin, Hojin and Lee, Yongmin and Yoo, Seehwan and Choi, Jongmoo},
        journal = {PVLDB},
        series = {{VLDB} '26},
        volume = {19},
        number = {11},
        pages = {3202--3215},
        doi = {10.14778/3836663.3836668},
        url = {https://doi.org/10.14778/3836663.3836668},
        year = {2026}
}

Incoming Citations (Sorted by Pagerank)

Showing 0 of 0 citing papers.

Rank Citing Paper Year Venue Pagerank
Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 17 of 17 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 0.00046284649
78 Automatic Database Management System Tuning Through Large-scale Machine Learning 2017 SIGMOD 0.00036684414
314 An End-to-End Automatic Cloud Database Tuning System Using Deep Reinforcement Learning 2019 SIGMOD 0.00021282642
400 Monkey: Optimal Navigable Key-Value Store 2017 SIGMOD 0.00019129175
430 ALEX: An Updatable Adaptive Learned Index 2020 SIGMOD 0.00018409112
437 QTune: A Query-Aware Database Tuning System with Deep Reinforcement Learning 2019 VLDB 0.00018315867
555 SageDB: A Learned Database System 2019 CIDR 0.00016506678
754 Dostoevsky: Better Space-Time Trade-Offs for LSM-Tree Based Key-Value Stores via Adaptive Removal of Superfluous Merging 2018 SIGMOD 0.00014236015
816 SlimDB: A Space-Efficient Key-Value Storage Engine For Semi-Sorted Data 2017 VLDB 0.00013687311
1,302 Apache Hadoop Goes Realtime at Facebook 2011 SIGMOD 0.00011111126
1,422 The Log-Structured Merge-Bush & the Wacky Continuum 2019 SIGMOD 0.00010725538
4,041 Disaggregating RocksDB: A Production Experience 2023 SIGMOD 6.8371576e-05
4,121 On Performance Stability in LSM-based Storage Systems 2020 VLDB 6.7946812e-05
5,248 Endure: A Robust Tuning Paradigm for LSM Trees Under Workload Uncertainty 2022 VLDB 6.211056e-05
6,124 Learning to Optimize LSM-trees: Towards A Reinforcement Learning based Key-Value Store for Dynamic Workloads 2023 SIGMOD 5.8788211e-05
7,241 CaaS-LSM: Compaction-as-a-Service for LSM-based Key-Value Stores in Storage Disaggregated Infrastructure 2024 SIGMOD 5.5777445e-05
8,221 ArceKV: Towards Workload-driven LSM-compactions for Key-Value Store Under Dynamic Workloads 2026 VLDB 5.3759492e-05
Previous Page 1 / 1 Next

Semantically Similar Papers