Comprehensive Evaluation of GNN Training Systems: A Data Management Perspective

Autor:	Yuan, Hao, Liu, Yajiong, Zhang, Yanfeng, Ai, Xin, Wang, Qiange, Chen, Chaoyi, Gu, Yu, Yu, Ge
Rok vydání:	2023
Předmět:	Computer Science - Machine Learning Computer Science - Distributed Parallel and Cluster Computing
Druh dokumentu:	Working Paper
Popis:	Many Graph Neural Network (GNN) training systems have emerged recently to support efficient GNN training. Since GNNs embody complex data dependencies between training samples, the training of GNNs should address distinct challenges different from DNN training in data management, such as data partitioning, batch preparation for mini-batch training, and data transferring between CPUs and GPUs. These factors, which take up a large proportion of training time, make data management in GNN training more significant. This paper reviews GNN training from a data management perspective and provides a comprehensive analysis and evaluation of the representative approaches. We conduct extensive experiments on various benchmark datasets and show many interesting and valuable results. We also provide some practical tips learned from these experiments, which are helpful for designing GNN training systems in the future. Comment: 12 pages, 18 figures. (Accepted by VLDB 2024)
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2311.13279 Zobrazit plný text záznamu View this record from Arxiv