EnGN, a simulated edge-centric GNN accelerator with ring-edge-reduce dataflow and degree-aware caching, claims 2.97x speedup and 6.2x energy efficiency over the HyGCN accelerator.
Imagenet classifi- cation with deep convolutional neural networks,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.DC 1years
2019 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
EnGN: A High-Throughput and Energy-Efficient Accelerator for Large Graph Neural Networks
EnGN, a simulated edge-centric GNN accelerator with ring-edge-reduce dataflow and degree-aware caching, claims 2.97x speedup and 6.2x energy efficiency over the HyGCN accelerator.