HetCache: Synergising NVMe Storage and GPU acceleration for Memory-Efficient Analytics

Summary: HetCache co-optimizes data placement and access paths across NVMe, host DRAM and GPU memory for heterogeneous CPU–GPU servers, treating bandwidth/costs as access-path- and device-dependent. Using proportional, access-path-aware caching, it yields 1.14–1.78× speedups on NVMe-resident workloads and near in-memory performance for hybrid CPU–GPU execution while greatly improving memory efficiency. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID: 497
Venue: CIDR
Year: 2023
Pagerank: 5.8651793e-05
Overall Rank: 6,487 | 55.50%
DOI: -

Incoming Non-self Citations Over Time

Authors

1. Hamish Nicholson (EPFL)
2. Aunn Raza (EPFL)
3. Periklis Chrysogelos (EPFL; Oracle)
4. Anastasia Ailamaki (EPFL)

BibTeX Citation

@inproceedings{nicholson_cidr23,
        address = {Amsterdam, Netherlands},
        series = {{CIDR} '23},
        title = {{HetCache: Synergising NVMe Storage and GPU acceleration for Memory-Efficient Analytics}},
        booktitle = {Proceedings of the {Conference} on {Innovative} {Data} {Systems} {Research}},
        author = {Nicholson, Hamish and Raza, Aunn and Chrysogelos, Periklis and Ailamaki, Anastasia},
        year = {2023}
}

Incoming Citations (Sorted by Pagerank)

Showing 6 of 6 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
5,778	GOLAP: A GPU-in-Data-Path Architecture for High-Speed OLAP	2024	SIGMOD	6.0918646e-05
7,432	Powerful GPUs or Fast Interconnects: Analyzing Relational Workloads on Modern GPUs	2025	VLDB	5.6199783e-05
8,808	Scaling your Hybrid CPU-GPU DBMS to Multiple GPUs	2024	VLDB	5.3661351e-05
9,873	Workload Placement on Heterogeneous CPU-GPU Systems	2024	VLDB	5.2043672e-05
10,127	Declarative Memory Services	2026	CIDR	5.093636e-05
11,200	High-Performance Query Processing with NVMe Arrays: Spilling without Killing Performance	2024	SIGMOD	5.093636e-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
135	An Evaluation of Buffer Management Strategies for Relational Database Systems	1985	VLDB	0.00030209257
422	Umbra: A Disk-Based System with In-Memory Performance	2020	CIDR	0.00018732744
637	2Q: A Low Overhead High Performance Buffer Management Replacement Algorithm	1994	VLDB	0.0001547341
823	The Yin and Yang of Processing Data Warehousing Queries on GPU Devices	2013	VLDB	0.00013792901
1,466	A Study of the Fundamental Performance Characteristics of GPUs and CPUs for Database Analytics	2020	SIGMOD	0.0001068941
1,490	SSD Bufferpool Extensions for Database Systems	2010	VLDB	0.00010595269
1,574	Pipelined Query Processing in Coprocessor Environments	2018	SIGMOD	0.00010321274
1,666	HippogriffDB: Balancing I/O and GPU Bandwidth in Big Data Analytics	2016	VLDB	0.00010068964
1,977	HetExchange: Encapsulating heterogeneous CPU-GPU parallelism in JIT compiled engines	2019	VLDB	9.3641101e-05
2,267	In-Memory Performance for Big Data	2015	VLDB	8.8388928e-05
2,558	The Oracle Universal Server Buffer Manager	1997	VLDB	8.41572e-05
2,828	Exploiting Directly-Attached NVMe Arrays in DBMS	2020	CIDR	8.0803679e-05
3,791	Hardware-conscious Query Processing in GPU-accelerated Analytical Engines	2019	CIDR	7.1235328e-05
6,835	GPU-accelerated data management under the test of time	2020	CIDR	5.7584924e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	3,791	Hardware-conscious Query Processing in GPU-accelerated Analytical Engines	2019	CIDR
2	3,761	ScaleStore: A Fast and Cost-Efficient Storage Engine using DRAM, NVMe, and RDMA	2022	SIGMOD
3	2,566	Pump Up the Volume: Processing Large Data on GPUs with Fast Interconnects	2020	SIGMOD
4	10,985	Scaling GPU-Accelerated Databases beyond GPU Memory Size	2025	VLDB
5	1,666	HippogriffDB: Balancing I/O and GPU Bandwidth in Big Data Analytics	2016	VLDB
6	10,901	STsCache: An Efficient Semantic Caching Scheme for Time-series Data Workloads Based on Hybrid Storage	2025	VLDB
7	4,243	Orchestrating Data Placement and Query Execution in Heterogeneous CPU-GPU DBMS	2022	VLDB
8	6,835	GPU-accelerated data management under the test of time	2020	CIDR
9	1,977	HetExchange: Encapsulating heterogeneous CPU-GPU parallelism in JIT compiled engines	2019	VLDB
10	4,764	ReCache: Reactive Caching for Fast Analytics over Heterogeneous Data	2018	VLDB