Themis: A GPU-accelerated Relational Query Execution Engine

Summary: Themis is a GPU relational engine addressing intra- and inter-warp imbalance from skewed, varying-cardinality pipelines. It redistributes tuples within and across warps using adaptive workload estimates, achieving up to 379× gains on skewed TPC-H. (summarized by gpt-5.6-luna on Jul 24 2026)

Paper ID: hc27442970ea8d016
Venue: VLDB
Year: 2025
Pagerank: 5.1861628e-05
Overall Rank: 9,392 | 36.86%
DOI: 10.14778/3705829.3705856

Incoming Non-self Citations Over Time

Authors

1. Kijae Hong (Pohang University of Science and Technology)
2. Kyoungmin Kim (EPFL)
3. Young-Koo Lee (Kyunghee University)
4. Yang-Sae Moon (Kangwon National University)
5. Sourav S Bhowmick (Nanyang Technological University)
6. Wook-Shin Han (Pohang University of Science and Technology)

BibTeX Citation

@article{hong_vldb25,
        title = {{Themis: A GPU-accelerated Relational Query Execution Engine}},
        author = {Hong, Kijae and Kim, Kyoungmin and Lee, Young-Koo and Moon, Yang-Sae and Bhowmick, Sourav S and Han, Wook-Shin},
        journal = {PVLDB},
        series = {{VLDB} '25},
        volume = {18},
        number = {2},
        pages = {426--438},
        doi = {10.14778/3705829.3705856},
        url = {https://doi.org/10.14778/3705829.3705856},
        year = {2025}
}

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
5,991	Terabyte-Scale Analytics in the Blink of an Eye	2026	VLDB	5.9247663e-05
9,863	GPU Acceleration of SQL Analytics on Compressed Data	2026	VLDB	5.1176637e-05

Outgoing Citations (Sorted by Pagerank)

Showing 16 of 16 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
215	Morsel-Driven Parallelism: A NUMA-Aware Query Evaluation Framework for the Many-Core Age	2014	SIGMOD	0.00024598661
616	Relational Joins on Graphics Processors	2008	SIGMOD	0.00015561564
1,268	A Study of the Fundamental Performance Characteristics of GPUs and CPUs for Database Analytics	2020	SIGMOD	0.0001126007
1,473	Voodoo - A Vector Algebra for Portable Database Performance on Modern Hardware	2016	VLDB	0.00010557973
1,533	Pipelined Query Processing in Coprocessor Environments	2018	SIGMOD	0.00010332035
1,603	SkinnerDB: Regret-Bounded Query Evaluation via Reinforcement Learning	2019	SIGMOD	0.00010097649
1,747	HetExchange: Encapsulating heterogeneous CPU-GPU parallelism in JIT compiled engines	2019	VLDB	9.7335416e-05
2,468	GPU-Accelerated Subgraph Enumeration on Partitioned Graphs	2020	SIGMOD	8.4178183e-05
2,815	GPL: A GPU-based Pipelined Query Processing Engine	2016	SIGMOD	7.9777435e-05
3,653	Realtime Top-k Personalized PageRank over Large Graphs on GPUs	2020	VLDB	7.1288161e-05
3,961	Data-Parallel Query Processing on Non-Uniform Data	2020	VLDB	6.8948667e-05
4,480	GPU Database Systems Characterization and Optimization	2024	VLDB	6.5807917e-05
4,645	Improving Execution Efficiency of Just-in-time Compilation based Query Processing on GPUs	2021	VLDB	6.4881339e-05
8,850	SkinnerMT: Parallelizing for Efficiency and Robustness in Adaptive Query Processing on Multicore Platforms	2023	VLDB	5.2645282e-05
8,851	Workload-driven, Lazy Discovery of Data Dependencies for Query Optimization	2022	CIDR	5.2644901e-05
9,839	Cost-efficiency and Performance Robustness in Serverless Data Exchange	2022	SIGMOD	5.1234085e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	2,815	GPL: A GPU-based Pipelined Query Processing Engine	2016	SIGMOD
2	2,600	Robust Query Processing in Co-Processor-accelerated Databases	2016	SIGMOD
3	6,587	Maximus: A Modular Accelerated Query Engine for Data Analytics on Heterogeneous Systems	2025	SIGMOD
4	3,888	Orchestrating Data Placement and Query Execution in Heterogeneous CPU-GPU DBMS	2022	VLDB
5	2,466	Concurrent Analytical Query Processing with GPUs	2014	VLDB
6	3,961	Data-Parallel Query Processing on Non-Uniform Data	2020	VLDB
7	10,601	TQEx: Tensor-based Query Engine Enhanced by Bridging the Gap	2026	SIGMOD
8	10,911	Hermes at Scale: Powering Distributed Queries with a Unified Memory Fabric	2026	VLDB
9	3,773	Hardware-conscious Query Processing in GPU-accelerated Analytical Engines	2019	CIDR
10	8,490	THEMIS: Fairness in Federated Stream Processing under Overload	2016	SIGMOD