most citedHongTu: Scalable Full-Graph GNN Training on Multiple GPUs (via communication-optimized CPU data offloading)

23 citations · 23 across the 1 of their papers we have counts for

collaborators

1 paper

cs.DC2023★ 23 cited

HongTu: Scalable Full-Graph GNN Training on Multiple GPUs (via communication-optimized CPU data offloading)

Qiange Wang, Yao Chen, Weng-Fai Wong +1

Full-graph training on graph neural networks (GNN) has emerged as a promising training method for its effectiveness. Full-graph training requires extensive memory and computation r…