DBScholar

Back to papers

SANCUS: Staleness-Aware Communication-Avoiding Full-Graph Decentralized Training in Large-Scale Graph Neural Networks

Summary: SANCUS is a decentralized full-graph GNN trainer that uses bounded embedding-staleness metrics, cached historical embeddings, and adaptive broadcast skipping to avoid communication. It provides convergence guarantees and achieves up to 74% less communication and 1.86× higher throughput without accuracy loss. (summarized by gpt-5.6-luna on Jul 24 2026)

Paper ID
h33e4d0ef20b4aa07
Venue
VLDB
Year
2022
Pagerank
0.00012313386
Overall Rank
1,047 | 92.99%
DOI
10.14778/3538598.3538614
PDF
Download (CC BY-NC-ND 4.0)

Incoming Non-self Citations Over Time

Authors

BibTeX Citation

@article{peng_vldb22,
        title = {{SANCUS: Staleness-Aware Communication-Avoiding Full-Graph Decentralized Training in Large-Scale Graph Neural Networks}},
        author = {Peng, Jingshu and Chen, Zhao and Shao, Yingxia and Shen, Yanyan and Chen, Lei and Cao, Jiannong},
        journal = {PVLDB},
        series = {{VLDB} '22},
        volume = {15},
        number = {9},
        pages = {1937--1950},
        doi = {10.14778/3538598.3538614},
        url = {https://doi.org/10.14778/3538598.3538614},
        year = {2022}
}

Incoming Citations (Sorted by Pagerank)

Showing 31 of 31 citing papers.

Rank Citing Paper Year Venue Pagerank
2,325 Scalable and Efficient Full-Graph GNN Training for Large Graphs 2023 SIGMOD 8.625677e-05
2,513 DUCATI: A Dual-Cache Training System for Graph Neural Networks on Giant Graphs with the GPU 2023 SIGMOD 8.3467757e-05
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
5,020 DGC: Training Dynamic Graphs with Spatio-Temporal Non-Uniformity using Graph Partitioning by Chunks 2023 SIGMOD 6.2996768e-05
5,332 DynaHB: A Communication-Avoiding Asynchronous Distributed Framework with Hybrid Batches for Dynamic GNN Training 2024 VLDB 6.1623722e-05
5,359 FreshGNN: Reducing Memory Access via Stable Historical Embeddings for Graph Neural Network Training 2024 VLDB 6.1498425e-05
5,916 Lotan: Bridging the Gap between GNNs and Scalable Graph Analytics Engines 2023 VLDB 5.9320442e-05
6,686 Efficient Training of Graph Neural Networks on Large Graphs 2024 VLDB 5.6931383e-05
6,849 SIMPLE: Efficient Temporal Graph Neural Network Training at Scale with Dynamic Data Placement 2024 SIGMOD 5.6488938e-05
6,947 HongTu: Scalable Full-Graph GNN Training on Multiple GPUs 2023 SIGMOD 5.6213095e-05
7,036 Can Graph Reordering Speed Up Graph Neural Network Training? An Experimental Study 2025 VLDB 5.5993909e-05
7,544 NeutronTask: Scalable and Efficient Multi-GPU GNN Training with Task Parallelism 2025 VLDB 5.4817498e-05
7,729 Systems for Scalable Graph Analytics and Machine Learning: Trends and Methods 2025 VLDB 5.4511876e-05
8,174 NeutronTP: Load-Balanced Distributed Full-Graph GNN Training with Tensor Parallelism 2025 VLDB 5.3692867e-05
8,649 D3-GNN: Dynamic Distributed Dataflow for Streaming Graph Neural Networks 2024 VLDB 5.2809021e-05
9,835 NeutronHeter: Optimizing Distributed Graph Neural Network Training for Heterogeneous Clusters 2026 SIGMOD 5.1064323e-05
9,942 Scalable Graph Convolutional Network Training on Distributed-Memory Systems 2023 VLDB 5.0923288e-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
10,404 NeutronSparse: Coordinating Heterogeneous Engines for Sparse Matrix Multiplication on NPUs 2027 SIGMOD 4.9605343e-05
10,777 Efficient GNN Training on Giant Graphs with Collective Batching and Scheduling 2026 VLDB 4.9605343e-05
10,808 FlareDTDG: Harnessing Temporal Recency for Scalable Discrete-Time Dynamic Graph Training 2026 VLDB 4.9605343e-05
11,113 NeutronCloud: Resource-Aware Distributed GNN Training in Fluctuating Cloud Environments 2026 VLDB 4.9605343e-05
11,266 SWASH: A Flexible Communication Framework with Sliding Window-Based Cache Sharing for Scalable DGNN Training 2025 SIGMOD 4.9605343e-05
11,352 Heta: Distributed Training of Heterogeneous Graph Neural Networks 2025 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 13 of 13 cited papers.

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

Rank Cited Paper Year Venue Pagerank
3 Pregel: A System for Large-Scale Graph Processing 2010 SIGMOD 0.0012089097
203 AliGraph: A Comprehensive Graph Neural Network Platform 2019 VLDB 0.00025234479
902 Parallelizing Sequential Graph Computations 2017 SIGMOD 0.00013142001
1,021 AGL: A Scalable System for Industrial-purpose Graph Machine Learning 2020 VLDB 0.00012408574
1,223 Large Graph Convolutional Network Training with GPU-Oriented Data Communication Architecture 2021 VLDB 0.00011439503
2,196 Heterogeneity-aware Distributed Parameter Servers 2017 SIGMOD 8.8807201e-05
2,434 Accelerating Large Scale Real-Time GNN Inference using Channel Pruning 2021 VLDB 8.4608438e-05
3,706 G3: When Graph Neural Networks Meet Parallel Graph Processing Systems on GPUs 2020 VLDB 7.0605656e-05
3,878 Grain: Improving Data Efficiency of Graph Neural Networks via Diversified Influence Maximization 2021 VLDB 6.9334452e-05
4,235 Accelerating Dynamic Graph Analytics on GPUs 2018 VLDB 6.7032322e-05
4,961 Heterogeneity-Aware Distributed Machine Learning Training via Partial Reduce 2021 SIGMOD 6.3246342e-05
6,151 GRAPE: Parallelizing Sequential Graph Computations 2017 VLDB 5.8553261e-05
6,693 Reliable Data Distillation on Graph Convolutional Network 2020 SIGMOD 5.6909845e-05
Previous Page 1 / 1 Next

Semantically Similar Papers

# Overall Rank Paper Year Venue
1 5,359 FreshGNN: Reducing Memory Access via Stable Historical Embeddings for Graph Neural Network Training 2024 VLDB
2 11,421 Faster Convergence in Mini-batch Graph Neural Networks Training with Pseudo Full Neighborhood Compensation 2025 VLDB
3 5,020 DGC: Training Dynamic Graphs with Spatio-Temporal Non-Uniformity using Graph Partitioning by Chunks 2023 SIGMOD
4 6,686 Efficient Training of Graph Neural Networks on Large Graphs 2024 VLDB
5 8,656 Historical Embedding-Guided Efficient Large-Scale Federated Graph Learning 2024 SIGMOD
6 1,780 ByteGNN: Efficient Graph Neural Network Training at Large Scale 2022 VLDB
7 9,942 Scalable Graph Convolutional Network Training on Distributed-Memory Systems 2023 VLDB
8 7,530 ADGNN: Towards Scalable GNN Training with Aggregation-Difference Aware Sampling 2023 SIGMOD
9 8,649 D3-GNN: Dynamic Distributed Dataflow for Streaming Graph Neural Networks 2024 VLDB
10 2,325 Scalable and Efficient Full-Graph GNN Training for Large Graphs 2023 SIGMOD