SUGAR: Efficient Subgraph-level Training via Resource-aware Graph Partitioning
This addresses resource constraints for GNN deployment in edge computing and IoT applications, representing an incremental improvement in efficiency.
The paper tackles the problem of resource-efficient training for Graph Neural Networks (GNNs) by proposing SUGAR, a method that partitions graphs into subgraphs for local training, achieving up to 33 times runtime speedup and 3.8 times memory reduction on large-scale graphs.
Graph Neural Networks (GNNs) have demonstrated a great potential in a variety of graph-based applications, such as recommender systems, drug discovery, and object recognition. Nevertheless, resource-efficient GNN learning is a rarely explored topic despite its many benefits for edge computing and Internet of Things (IoT) applications. To improve this state of affairs, this work proposes efficient subgraph-level training via resource-aware graph partitioning (SUGAR). SUGAR first partitions the initial graph into a set of disjoint subgraphs and then performs local training at the subgraph-level. We provide a theoretical analysis and conduct extensive experiments on five graph benchmarks to verify its efficacy in practice. Our results show that SUGAR can achieve up to 33 times runtime speedup and 3.8 times memory reduction on large-scale graphs. We believe SUGAR opens a new research direction towards developing GNN methods that are resource-efficient, hence suitable for IoT deployment.