H-Rocks: CPU-GPU accelerated Heterogeneous RocksDB on Persistent Memory

Summary: H-Rocks merges CPU-GPU to accelerate RocksDB on persistent memory, offloading hot-paths to the GPU while preserving compatibility via key-value versioning. Uses sub-batching and minimizes CPU-GPU data movement, delivering 3–18x YCSB gains over CPU-based pKVS. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: haf60744d60897040
Venue: SIGMOD
Year: 2025
Pagerank: 5.1571823e-05
Overall Rank: 9,562 | 35.72%
DOI: 10.1145/3709694

Incoming Non-self Citations Over Time

Authors

1. Shweta Pandey (Indian Institute of Science)
2. Arkaprava Basu (Indian Institute of Science)

BibTeX Citation

@inproceedings{pandey_sigmod25,
        title = {{H-Rocks: CPU-GPU accelerated Heterogeneous RocksDB on Persistent Memory}},
        author = {Pandey, Shweta and Basu, Arkaprava},
        series = {{SIGMOD} '25},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3709694},
        url = {https://dl.acm.org/doi/10.1145/3709694},
        year = {2025}
}

Incoming Citations (Sorted by Pagerank)

Showing 1 of 1 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
11,278	BLAEQ: A Multigrid Index for Spatial Query on Geometry Data	2025	VLDB	4.9793485e-05

Outgoing Citations (Sorted by Pagerank)

Showing 14 of 14 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
502	FPTree: A Hybrid SCM-DRAM Persistent and Concurrent B-Tree for Storage Class Memory	2016	SIGMOD	0.00017250534
546	Faster: A Concurrent Key-Value Store with In-Place Updates	2018	SIGMOD	0.00016590738
1,147	MyRocks: LSM-Tree Database Storage Engine Serving Facebook's Social Graph	2020	VLDB	0.00011813751
1,359	Dash: Scalable Hashing on Persistent Memory	2020	VLDB	0.00010919021
2,202	Mega-KV: A Case for GPUs to Maximize the Throughput of In-Memory Key-Value Stores	2015	VLDB	8.8719608e-05
2,274	DPTree: Differential Indexing for Persistent Memory	2020	VLDB	8.7108312e-05
2,284	Pump Up the Volume: Processing Large Data on GPUs with Fast Interconnects	2020	SIGMOD	8.6954168e-05
2,510	Viper: An Efficient Hybrid PMem-DRAM Key-Value Store	2021	VLDB	8.3698208e-05
3,888	Orchestrating Data Placement and Query Execution in Heterogeneous CPU-GPU DBMS	2022	VLDB	6.945725e-05
4,041	Disaggregating RocksDB: A Production Experience	2023	SIGMOD	6.8371576e-05
5,483	Plush: A Write-Optimized Persistent Log-Structured Hash-Table	2022	VLDB	6.1103271e-05
7,760	IcebergHT: High Performance Hash Tables Through Stability and Low Associativity	2023	SIGMOD	5.457277e-05
9,690	FluidKV: Seamlessly Bridging the Gap between Indexing Performance and Memory-Footprint on Ultra-Fast Storage	2024	VLDB	5.1405104e-05
9,691	BonsaiKV: Towards Fast, Scalable, and Persistent Key-Value Stores with Tiered, Heterogeneous Memory System	2024	VLDB	5.1405104e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	8,274	Scaling GPU-Accelerated Databases beyond GPU Memory Size	2025	VLDB
2	9,081	MirrorKV: An Efficient Key-Value Store on Hybrid Cloud Storage with Balanced Performance of Compaction and Querying	2023	SIGMOD
3	6,489	HetCache: Synergising NVMe Storage and GPU acceleration for Memory-Efficient Analytics	2023	CIDR
4	1,541	HippogriffDB: Balancing I/O and GPU Bandwidth in Big Data Analytics	2016	VLDB
5	8,796	GPH: An Efficient and Effective Perfect Hashing Scheme for GPU Architectures	2025	SIGMOD
6	10,933	SOS: A High-Performance Distributed Key-Value Store for Large-Scale Online Services	2026	VLDB
7	2,600	Robust Query Processing in Co-Processor-accelerated Databases	2016	SIGMOD
8	2,202	Mega-KV: A Case for GPUs to Maximize the Throughput of In-Memory Key-Value Stores	2015	VLDB
9	10,856	How Much Can RocksDB Chew? Achieving Near-Zero Write Stalls with Sustainable RocksDB	2026	VLDB
10	436	Optimizing Space Amplification in RocksDB	2017	CIDR