mlsanityData to 7 Oct 2026

EnGN: A High-Throughput and Energy-Efficient Accelerator for Large Graph Neural Networks

arXiv 1909.00155 ↗First posted 31 Aug 2019cs.DCcited by 42

Not on the map. Why?

Abstract (arXiv)

Graph neural networks (GNNs) emerge as a powerful approach to process non-euclidean data structures and have been proved powerful in various application domains such as social networks and e-commerce. While such graph data maintained in real-world systems can be extremely large and sparse, thus employing GNNs to deal with them requires substantial computational and memory overhead, which induces considerable energy and resource cost on CPUs and GPUs. In this work, we present a specialized accelerator architecture, EnGN, to enable high-throughput and energy-efficient processing of large-scale GNNs. The proposed EnGN is designed to accelerate the three key stages of GNN propagation, which is abstracted as common computing patterns shared by typical GNNs. To support the key stages simultaneously, we propose the ring-edge-reduce(RER) dataflow that tames the poor locality of sparsely-and-randomly connected vertices, and the RER PE-array to practice RER dataflow. In addition, we utilize a graph tiling strategy to fit large graphs into EnGN and make good use of the hierarchical on-chip buffers through adaptive computation reordering and tile scheduling. Overall, EnGN achieves performance speedup by 1802.9X, 19.75X, and 2.97X and energy efficiency by 1326.35X, 304.43X, and 6.2X on average compared to CPU, GPU, and a state-of-the-art GCN accelerator HyGCN, respectively.

Key facts. Cited by 42 papers in the map (data through 7 Oct 2026). First posted on arXiv 31 Aug 2019.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 9; 2022: 12; 2023: 9; 2024: 8; 2025: 4.

42 papers.

Sort:

  1. 2022Trustworthy Graph Neural Networks: Aspects, Methods and Trendscited by 88
  2. 2021Sampling methods for efficient training of graph convolutional networks: A surveycited by 40
  3. 2022Parallel and Distributed Graph Neural Networks: An In-Depth Concurrency Analysiscited by 29
  4. 2023MaxK-GNN: Extremely Fast GPU Kernel Design for Accelerating Graph Neural Networks Trainingcited by 25
  5. 2022I-GCN: A Graph Convolutional Network Accelerator with Runtime Locality Enhancement through Islandizationcited by 21
  6. 2023A Survey on Graph Neural Network Acceleration: Algorithms, Systems, and Customized Hardwarecited by 18
  7. 2021G-CoS: GNN-Accelerator Co-Search Towards Both Better Accuracy and Efficiencycited by 15
  8. 2021GCoD: Graph Convolutional Network Acceleration via Dedicated Algorithm and Accelerator Co-Designcited by 12
  9. 2022SmartSAGE: Training Large-scale Graph Neural Networks using In-Storage Processing Architecturescited by 12
  10. 2022LL-GNN: Low Latency Graph Neural Networks on FPGAs for High Energy Physicscited by 12
  11. 2021BlockGNN: Towards Efficient GNN Acceleration Using Block-Circulant Weight Matricescited by 11
  12. 2022Neural Graph Databasescited by 11
30 more
  1. 2022FlowGNN: A Dataflow Architecture for Real-Time Workload-Agnostic Graph Neural Network Inferencecited by 10
  2. 2023HOT: Higher-Order Dynamic Graph Representation Learning with Efficient Transformerscited by 10
  3. 2022GenGNN: A Generic FPGA Framework for Graph Neural Network Accelerationcited by 9
  4. 2023GHOST: A Graph Neural Network Accelerator using Silicon Photonicscited by 7
  5. 2021GNNear: Accelerating Full-Batch Training of Graph Neural Networks with Near-Memory Processingcited by 6
  6. 2022GROW: A Row-Stationary Sparse-Dense GEMM Accelerator for Memory-Efficient Graph Convolutional Neural Networkscited by 4
  7. 2023SGCN: Exploiting Compressed-Sparse Features in Deep Graph Convolutional Network Acceleratorscited by 4
  8. 2021VersaGNN: a Versatile accelerator for Graph neural networkscited by 3
  9. 2021GNNIE: GNN Inference Engine with Load-balancing and Graph-Specific Cachingcited by 3
  10. 2021LW-GCN: A Lightweight FPGA-based Graph Convolutional Network Acceleratorcited by 2
  11. 2024PyGim: An Efficient Graph Neural Network Library for Real Processing-In-Memory Architecturescited by 2
  12. 2024Edge Graph Intelligence: Reciprocally Empowering Edge Networks with Graph Intelligencecited by 2
  13. 2025GraNNite: Enabling High-Performance Execution of Graph Neural Networks on Resource-Constrained Neural Processing Unitscited by 2
  14. 2022Hardware/Software Co-Programmable Framework for Computational SSDs to Accelerate Deep Learning Service on Large-Scale Graphscited by 1
  15. 2022H-GCN: A Graph Convolutional Network Accelerator on Versal ACAP Architecturecited by 1
  16. 2023Slice-and-Forge: Making Better Use of Caches for Graph Convolutional Network Acceleratorscited by 1
  17. 2023GNNBuilder: An Automated Framework for Generic Graph Neural Network Accelerator Generation, Simulation, and Optimizationcited by 1
  18. 2023Low Latency Edge Classification GNN for Particle Trajectory Tracking on FPGAscited by 1
  19. 2024Accelerating Neural Networks for Large Language Models and Graph Processing with Silicon Photonicscited by 1
  20. 2024Single-GPU GNN Systems: Traps and Pitfallscited by 1
  21. 2024Optimization of GNN Training Through Half-precisioncited by 1
  22. 2025ADAPT: Lightweight, Long-Range Machine Learning Force Fields Without Graphscited by 1
  23. 2021SPA-GCN: Efficient and Flexible GCN Accelerator with an Application for Graph Similarity Computationcited by 0
  24. 2022COIN: Communication-Aware In-Memory Acceleration for Graph Convolutional Networkscited by 0
  25. 2023Accelerating Generic Graph Neural Networks via Architecture, Compiler, Partition Method Co-Designcited by 0
  26. 2024NeuraChip: Accelerating GNN Computations with a Hash-based Decoupled Spatial Acceleratorcited by 0
  27. 2024Acceleration Algorithms in GNNs: A Surveycited by 0
  28. 2024Efficient Message Passing Architecture for GCN Training on HBM-based FPGAs with Orthogonal Topology On-Chip Networkscited by 0
  29. 2025MixGCN: Scalable GCN Training by Mixture of Parallelism and Mixture of Acceleratorscited by 0
  30. 2025HePGA: A Heterogeneous Processing-in-Memory based GNN Training Acceleratorcited by 0