Distributed GPU Joins on Fast RDMA-capable Networks

Summary: Pipelined distributed GPU joins on fast RDMA networks overlap shuffling with build/probe to hide GPU idle time. RDMA/GPUDirect-based algorithms scale to arbitrarily large tables and show up to 6x faster full queries versus CPU-only joins. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: h7c3e52b7b5656740
Venue: SIGMOD
Year: 2023
Pagerank: 6.1584802e-05
Overall Rank: 5,371 | 63.89%
DOI: 10.1145/3588709

Incoming Non-self Citations Over Time

Authors

1. Lasse Thostrup (Technical University of Darmstadt)
2. Gloria Doci (Snowflake)
3. Nils Boeschen (Technical University of Darmstadt)
4. Manisha Luthra (German National Research Center for Information Technology; Technical University of Darmstadt)
5. Carsten Binnig (German National Research Center for Information Technology; Technical University of Darmstadt)

BibTeX Citation

@inproceedings{thostrup_sigmod23,
        title = {{Distributed GPU Joins on Fast RDMA-capable Networks}},
        author = {Thostrup, Lasse and Doci, Gloria and Boeschen, Nils and Luthra, Manisha and Binnig, Carsten},
        series = {{SIGMOD} '23},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3588709},
        url = {https://dl.acm.org/doi/10.1145/3588709},
        year = {2023}
}

Incoming Citations (Sorted by Pagerank)

Showing 14 of 14 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
5,991	Terabyte-Scale Analytics in the Blink of an Eye	2026	VLDB	5.9247663e-05
6,151	Efficiently Processing Joins and Grouped Aggregations on GPUs	2025	SIGMOD	5.8697699e-05
6,667	Powerful GPUs or Fast Interconnects: Analyzing Relational Workloads on Modern GPUs	2025	VLDB	5.7152948e-05
8,236	Scaling your Hybrid CPU-GPU DBMS to Multiple GPUs	2024	VLDB	5.3710413e-05
8,274	Scaling GPU-Accelerated Databases beyond GPU Memory Size	2025	VLDB	5.3642256e-05
8,752	Zero-sided RDMA: Network-driven Data Shuffling for Disaggregated Heterogeneous Cloud DBMSs	2024	SIGMOD	5.2847178e-05
9,429	Efficiently Joining Large Relations on Multi-GPU Systems	2025	VLDB	5.1786456e-05
9,687	DPDPU: Data Processing with DPUs	2025	CIDR	5.1408809e-05
10,621	Beluga: A CXL-Based Memory Architecture for Scalable and Efficient LLM KVCache Management	2026	SIGMOD	4.9793485e-05
10,723	Scalable GPU Acceleration of Scalar Functions in Analytical Databases: Compilation, Benchmarking, and Optimization	2026	VLDB	4.9793485e-05
10,785	MGI: A Communication Framework for Data Processing in Massive GPU Infrastructures	2026	VLDB	4.9793485e-05
10,801	PystachIO: Efficient Distributed GPU Query Processing with PyTorch over Fast Networks & Fast Storage	2026	VLDB	4.9793485e-05
11,436	Analyzing Near-Network Hardware Acceleration with Co-Processing on DPUs	2025	VLDB	4.9793485e-05
11,536	Enabling Adaptive Sampling for Intra-Window Join: Simultaneously Optimizing Quantity and Quality	2024	SIGMOD	4.9793485e-05

Outgoing Citations (Sorted by Pagerank)

Showing 20 of 20 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
251	Multi-Core, Main-Memory Joins: Sort vs. Hash Revisited	2014	VLDB	0.00023143736
616	Relational Joins on Graphics Processors	2008	SIGMOD	0.00015561564
661	Fast Sort on CPUs and GPUs: A Case for Bandwidth Oblivious SIMD Sort	2010	SIGMOD	0.00015003815
771	The Yin and Yang of Processing Data Warehousing Queries on GPU Devices	2013	VLDB	0.00014085862
889	Rack-Scale In-Memory Join Processing using RDMA	2015	SIGMOD	0.00013247362
945	The End of Slow Networks: It's Time for a Redesign	2016	VLDB	0.00012939225
1,268	A Study of the Fundamental Performance Characteristics of GPUs and CPUs for Database Analytics	2020	SIGMOD	0.0001126007
1,339	The End of a Myth: Distributed Transactions Can Scale	2017	VLDB	0.00010975458
1,366	Designing Distributed Tree-based Index Structures for Fast RDMA-capable Networks	2019	SIGMOD	0.00010912114
1,533	Pipelined Query Processing in Coprocessor Environments	2018	SIGMOD	0.00010332035
1,952	Quantifying TPC-H Choke Points and Their Optimizations	2020	VLDB	9.3189525e-05
2,122	Revisiting Co-Processing for Hash Joins on the Coupled CPU-GPU Architecture	2013	VLDB	9.0084047e-05
2,284	Pump Up the Volume: Processing Large Data on GPUs with Fast Interconnects	2020	SIGMOD	8.6954168e-05
2,600	Robust Query Processing in Co-Processor-accelerated Databases	2016	SIGMOD	8.2363864e-05
2,802	Distributed Join Algorithms on Thousands of Cores	2017	VLDB	7.9903139e-05
2,886	Efficient Join Algorithms For Large Database Tables in a Multi-GPU Environment	2021	VLDB	7.9081605e-05
3,053	MG-Join: A Scalable Join for Massively Parallel Multi-GPU Architectures	2021	SIGMOD	7.7019663e-05
3,200	Why it is time for a HyPE: A Hybrid Query Processing Engine for Efficient GPU Coprocessing in DBMS	2013	VLDB	7.5437496e-05
3,608	DFI: The Data Flow Interface for High-Speed Networks	2021	SIGMOD	7.1650501e-05
3,626	Triton Join: Efficiently Scaling to a Large Join State on GPUs with Fast Interconnects	2022	SIGMOD	7.1524537e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	3,059	Design and Evaluation of Parallel Pipelined Join Algorithms	1987	SIGMOD
2	2,122	Revisiting Co-Processing for Hash Joins on the Coupled CPU-GPU Architecture	2013	VLDB
3	6,516	GPU-accelerated data management under the test of time	2020	CIDR
4	2,284	Pump Up the Volume: Processing Large Data on GPUs with Fast Interconnects	2020	SIGMOD
5	3,626	Triton Join: Efficiently Scaling to a Large Join State on GPUs with Fast Interconnects	2022	SIGMOD
6	616	Relational Joins on Graphics Processors	2008	SIGMOD
7	889	Rack-Scale In-Memory Join Processing using RDMA	2015	SIGMOD
8	6,151	Efficiently Processing Joins and Grouped Aggregations on GPUs	2025	SIGMOD
9	9,429	Efficiently Joining Large Relations on Multi-GPU Systems	2025	VLDB
10	2,886	Efficient Join Algorithms For Large Database Tables in a Multi-GPU Environment	2021	VLDB