LPS-GNN : Deploying Graph Neural Networks on Graphs with 100-Billion Edges
This addresses the computational and memory bottlenecks in deploying GNNs on massive graphs for real-world applications like online platforms, though it is incremental in improving existing methods.
The paper tackles the challenge of scaling Graph Neural Networks to graphs with 100 billion edges by introducing LPS-GNN, a framework that achieves a 13.8% improvement in User Acquisition scenarios and can process such graphs in 10 hours on a single GPU.
Graph Neural Networks (GNNs) have emerged as powerful tools for various graph mining tasks, yet existing scalable solutions often struggle to balance execution efficiency with prediction accuracy. These difficulties stem from iterative message-passing techniques, which place significant computational demands and require extensive GPU memory, particularly when dealing with the neighbor explosion issue inherent in large-scale graphs. This paper introduces a scalable, low-cost, flexible, and efficient GNN framework called LPS-GNN, which can perform representation learning on 100 billion graphs with a single GPU in 10 hours and shows a 13.8% improvement in User Acquisition scenarios. We examine existing graph partitioning methods and design a superior graph partition algorithm named LPMetis. In particular, LPMetis outperforms current state-of-the-art (SOTA) approaches on various evaluation metrics. In addition, our paper proposes a subgraph augmentation strategy to enhance the model's predictive performance. It exhibits excellent compatibility, allowing the entire framework to accommodate various GNN algorithms. Successfully deployed on the Tencent platform, LPS-GNN has been tested on public and real-world datasets, achieving performance lifts of 8. 24% to 13. 89% over SOTA models in online applications.