WiseGraph: Optimizing GNN with Joint Workload Partition of Graph and Operations
Kezhao Huang, Jidong Zhai, Liyan Zheng, Haojie Wang, Yuyang Jin, Qihao Zhang, R.‐H. Zhang, Zhen Zheng, Youngmin Yi, Xipeng Shen
Abstract
Graph Neural Network (GNN) has emerged as an important workload for learning on graphs. With the size of graph data and the complexity of GNN model architectures increasing, developing an efficient GNN system grows more important. As GNN has heavy neural computation workloads on a large graph, it is crucial to partition the entire workload into smaller parts for parallel execution and optimization. However, existing approaches separately partition graph data and GNN operations, resulting in inefficiency and large data movement overhead.
Topics & Concepts
Computer scienceWorkloadPartition (number theory)Graph partitionInefficiencyComputationGraphParallel computingTheoretical computer scienceDistributed computingAlgorithmMathematicsOperating systemCombinatoricsEconomicsMicroeconomicsAdvanced Graph Neural NetworksGraph Theory and AlgorithmsMachine Learning in Materials Science