LavaStore: ByteDance's Purpose-built, High-performance, Cost-effective Local Storage Engine for Cloud Services

Summary: LavaStore tailors RocksDB for ByteDance’s write-heavy, cost- and tail-latency-sensitive workloads with KV separation, WAL specialization, and a user-space append-only filesystem. Deployed at >100 PB scale, it improves throughput, latency, and cost. (summarized by gpt-5.6-luna on Jul 24 2026)

Paper ID: 13772
Venue: VLDB
Year: 2024
Pagerank: 5.093636e-05
Overall Rank: 11,281 | 22.61%
DOI: 10.14778/3685800.3685807

Incoming Non-self Citations Over Time

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

Authors

1. Hao Wang (ByteDance)
2. Jiaxin Ou (ByteDance)
3. Ming Zhao (ByteDance)
4. Sheng Qiu (ByteDance)
5. Yizheng Jiao (ByteDance)
6. Yi Wang (ByteDance)
7. Qizhong Mao (ByteDance)
8. Zhengyu Yang (ByteDance)
9. Yang Liu (ByteDance)
10. Jianshun Zhang (ByteDance)
11. Jianyang Hu (ByteDance)
12. Jingwei Zhang (ByteDance)
13. Jinrui Liu (ByteDance)
14. Jiaqiang Chen (ByteDance)
15. Yong Shen (ByteDance)
16. Lixun Cao (ByteDance)
17. Heng Zhang (ByteDance)
18. Hongde Li (ByteDance)
19. Ming Li (ByteDance)
20. Yue Ma (ByteDance)
21. Lei Zhang (ByteDance)
22. Jian Liu (ByteDance)
23. Guanghui Zhang (ByteDance)
24. Fei Liu (ByteDance)
25. Jianjun Chen (ByteDance)

BibTeX Citation

@article{wang_vldb24,
        title = {{LavaStore: ByteDance's Purpose-built, High-performance, Cost-effective Local Storage Engine for Cloud Services}},
        author = {Wang, Hao and Ou, Jiaxin and Zhao, Ming and Qiu, Sheng and Jiao, Yizheng and Wang, Yi and Mao, Qizhong and Yang, Zhengyu and Liu, Yang and Zhang, Jianshun and Hu, Jianyang and Zhang, Jingwei and Liu, Jinrui and Chen, Jiaqiang and Shen, Yong and Cao, Lixun and Zhang, Heng and Li, Hongde and Li, Ming and Ma, Yue and Zhang, Lei and Liu, Jian and Zhang, Guanghui and Liu, Fei and Chen, Jianjun},
        journal = {PVLDB},
        series = {{VLDB} '24},
        volume = {17},
        number = {12},
        pages = {3799--3812},
        doi = {10.14778/3685800.3685807},
        url = {https://doi.org/10.14778/3685800.3685807},
        year = {2024}
}

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
10,616	Terark-DS: A High-Performance and Storage-Efficient Key-Value Separation Storage Engine on Disaggregated Storage	2026	VLDB	5.093636e-05
10,686	ABase: the Multi-Tenant NoSQL Serverless Database for Diverse and Dynamic Workloads in Large-scale Cloud Environments	2025	SIGMOD	5.093636e-05

Outgoing Citations (Sorted by Pagerank)

Showing 18 of 18 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
446	Monkey: Optimal Navigable Key-Value Store	2017	SIGMOD	0.00018332392
831	Dostoevsky: Better Space-Time Trade-Offs for LSM-Tree Based Key-Value Stores via Adaptive Removal of Superfluous Merging	2018	SIGMOD	0.00013748512
841	Let’s Talk About Storage & Recovery Methods for Non-Volatile Memory Database Systems	2015	SIGMOD	0.00013677356
857	SlimDB: A Space-Efficient Key-Value Storage Engine For Semi-Sorted Data	2017	VLDB	0.00013568886
880	SuRF: Practical Range Query Filtering with Fast Succinct Tries	2018	SIGMOD	0.00013432693
882	HOT: A Height Optimized Trie Index for Main-Memory Database Systems	2018	SIGMOD	0.0001342403
1,954	Chucky: A Succinct Cuckoo Filter for LSM-Tree	2021	SIGMOD	9.4208165e-05
2,471	Viper: An Efficient Hybrid PMem-DRAM Key-Value Store	2021	VLDB	8.5329393e-05
2,698	Rosetta: A Robust Space-Time Optimized Range Filter for Key-Value Stores	2020	SIGMOD	8.2450522e-05
2,919	SNARF: A Learning-Enhanced Range Filter	2022	VLDB	7.9628657e-05
3,001	What Modern NVMe Storage Can Do, And How To Exploit It: High-Performance I/O for High-Performance Storage Engines	2023	VLDB	7.8673493e-05
3,054	Spooky: Granulating LSM-Tree Compactions Correctly	2022	VLDB	7.8090808e-05
3,761	ScaleStore: A Fast and Cost-Efficient Storage Engine using DRAM, NVMe, and RDMA	2022	SIGMOD	7.1450129e-05
4,119	Proteus: A Self-Designing Range Filter	2022	SIGMOD	6.8907395e-05
4,888	Enabling Low Tail Latency on Multicore Key-Value Stores	2020	VLDB	6.4609843e-05
5,155	ByteHTAP: ByteDance’s HTAP System with High Data Freshness and Strong Data Consistency	2022	VLDB	6.3420293e-05
5,229	Dissecting, Designing, and Optimizing LSM-based Data Stores	2022	SIGMOD	6.307514e-05
6,792	Border-Collie: A Wait-free, Read-optimal Algorithm for Database Logging on Multicore Hardware	2019	SIGMOD	5.7721543e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	472	Optimizing Space Amplification in RocksDB	2017	CIDR
2	8,240	Magma: A High Data Density Storage Engine Used in Couchbase	2022	VLDB
3	4,200	Disaggregating RocksDB: A Production Experience	2023	SIGMOD
4	4,993	Key-Value Storage Engines	2020	SIGMOD
5	9,950	CloudJump: Optimizing Cloud Databases for Cloud Storages	2022	VLDB
6	8,919	MirrorKV: An Efficient Key-Value Store on Hybrid Cloud Storage with Balanced Performance of Compaction and Querying	2023	SIGMOD
7	4,632	LogStore: A Cloud-Native and Multi-Tenant Log Database	2021	SIGMOD
8	11,766	From Worst-Case to Average-Case Analysis: Accurate Latency Predictions for Key-Value Storage Engines	2020	SIGMOD
9	7,693	CaaS-LSM: Compaction-as-a-Service for LSM-based Key-Value Stores in Storage Disaggregated Infrastructure	2024	SIGMOD
10	10,471	Making LSM-Tree-based Key-Value Store Practical and Efficient for Multi-Tenant Serverless Cloud Databases	2026	SIGMOD