Artifact and instructions to generate experimental results for the Euro-Par 2023 proceedings paper: Auto-Divide GNN: Accelerating GNN Training with Subgraph Division
posted on 2024-05-15, 09:52authored byHongyu Chen, Zhejiang Ran, Keshi Ge, Zhiquan Lai, Jingfei Jiang, Dongsheng Li
Auto-Divide GNN accelerates GNN training by combining GPU caching and subgraph division. And it provides an auto-profiling method to achieve the optimal performance.