DBScholar

Back to papers

DUCATI: A Dual-Cache Training System for Graph Neural Networks on Giant Graphs with the GPU

Summary: DUCATI introduces a Dual-Cache system for GNNs on giant graphs, adding an Adj-Cache to exploit adjacency locality and accelerate mini-batch generation on GPUs. A workload-aware allocator tunes cache allocation; on 4B graphs, yields up to 3.3x speedups over DGL, with time–accuracy trade-offs analyzed. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID
hdc343db641070cf0
Venue
SIGMOD
Year
2023
Pagerank
8.3467757e-05
Overall Rank
2,513 | 83.16%
DOI
10.1145/3589311

Incoming Non-self Citations Over Time

Authors

BibTeX Citation

@inproceedings{zhang_sigmod23,
        title = {{DUCATI: A Dual-Cache Training System for Graph Neural Networks on Giant Graphs with the GPU}},
        author = {Zhang, Xin and Shen, Yanyan and Shao, Yingxia and Chen, Lei},
        series = {{SIGMOD} '23},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3589311},
        url = {https://dl.acm.org/doi/10.1145/3589311},
        year = {2023}
}

Incoming Citations (Sorted by Pagerank)

Showing 18 of 18 citing papers.

Rank Citing Paper Year Venue Pagerank
4,524 NeutronStream: A Dynamic GNN Training Framework with Sliding Window for Graph Streams 2024 VLDB 6.5492325e-05
4,785 ETC: Efficient Training of Temporal Graph Neural Networks over Large-scale Dynamic Graphs 2024 VLDB 6.4033344e-05
4,850 DAHA: Accelerating GNN Training with Data and Hardware Aware Execution Planning 2024 VLDB 6.3682022e-05
4,875 Comprehensive Evaluation of GNN Training Systems: A Data Management Perspective 2024 VLDB 6.3576702e-05
4,884 NeutronOrch: Rethinking Sample-based GNN Training under CPU-GPU Heterogeneous Environments 2024 VLDB 6.3564568e-05
4,910 Apt-Serve: Adaptive Request Scheduling on Hybrid Cache for Scalable LLM Inference Serving 2025 SIGMOD 6.3468554e-05
6,686 Efficient Training of Graph Neural Networks on Large Graphs 2024 VLDB 5.6931383e-05
6,801 Eliminating Data Processing Bottlenecks in GNN Training over Large Graphs via Two-level Feature Compression 2024 VLDB 5.6623697e-05
6,849 SIMPLE: Efficient Temporal Graph Neural Network Training at Scale with Dynamic Data Placement 2024 SIGMOD 5.6488938e-05
7,530 ADGNN: Towards Scalable GNN Training with Aggregation-Difference Aware Sampling 2023 SIGMOD 5.4862282e-05
7,544 NeutronTask: Scalable and Efficient Multi-GPU GNN Training with Task Parallelism 2025 VLDB 5.4817498e-05
9,835 NeutronHeter: Optimizing Distributed Graph Neural Network Training for Heterogeneous Clusters 2026 SIGMOD 5.1064323e-05
10,017 The Image Calculator: 10x Faster Image-AI Inference by Replacing JPEG with Self-designing Storage Format 2024 SIGMOD 5.0825892e-05
10,292 Graph Neural Network Training Systems: A Performance Comparison of Full-Graph and Mini-Batch 2025 VLDB 5.031399e-05
10,645 Gem: Scalable Monotonic Graph Processing Beyond Billion-Scale on a Single Machine 2026 SIGMOD 4.9605343e-05
10,777 Efficient GNN Training on Giant Graphs with Collective Batching and Scheduling 2026 VLDB 4.9605343e-05
10,889 ThunderGNN: Unlocking Tensor Cores for Graph Neural Networks 2026 VLDB 4.9605343e-05
11,421 Faster Convergence in Mini-batch Graph Neural Networks Training with Pseudo Full Neighborhood Compensation 2025 VLDB 4.9605343e-05
Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 3 of 3 cited papers.

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

Rank Cited Paper Year Venue Pagerank
392 One Trillion Edges: Graph Processing at Facebook-Scale 2015 VLDB 0.0001917435
1,047 SANCUS: Staleness-Aware Communication-Avoiding Full-Graph Decentralized Training in Large-Scale Graph Neural Networks 2022 VLDB 0.00012313386
1,223 Large Graph Convolutional Network Training with GPU-Oriented Data Communication Architecture 2021 VLDB 0.00011439503
Previous Page 1 / 1 Next

Semantically Similar Papers

# Overall Rank Paper Year Venue
1 5,020 DGC: Training Dynamic Graphs with Spatio-Temporal Non-Uniformity using Graph Partitioning by Chunks 2023 SIGMOD
2 5,332 DynaHB: A Communication-Avoiding Asynchronous Distributed Framework with Hybrid Batches for Dynamic GNN Training 2024 VLDB
3 2,325 Scalable and Efficient Full-Graph GNN Training for Large Graphs 2023 SIGMOD
4 6,815 XGNN: Boosting Multi-GPU GNN Training via Global GNN Memory Store 2024 VLDB
5 6,947 HongTu: Scalable Full-Graph GNN Training on Multiple GPUs 2023 SIGMOD
6 5,359 FreshGNN: Reducing Memory Access via Stable Historical Embeddings for Graph Neural Network Training 2024 VLDB
7 1,223 Large Graph Convolutional Network Training with GPU-Oriented Data Communication Architecture 2021 VLDB
8 4,848 Accelerating Sampling and Aggregation Operations in GNN Frameworks with GPU Initiated Direct Storage Accesses 2024 VLDB
9 6,801 Eliminating Data Processing Bottlenecks in GNN Training over Large Graphs via Two-level Feature Compression 2024 VLDB
10 10,777 Efficient GNN Training on Giant Graphs with Collective Batching and Scheduling 2026 VLDB