XGNN: Boosting Multi-GPU GNN Training via Global GNN Memory Store
Proceedings of the VLDB Endowment · 2024-01 · 17 (5) · 1105-1118
DOI: 10.14778/3641204.3641219 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the VLDB Endowment, online erschienen am 02.05.2024. Verfasst von Dahai Tang und 6 weiteren Autor:innen. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 57 Literaturangaben.
Abstract (Originalsprache)
GPUs are commonly utilized to accelerate GNN training, particularly on a multi-GPU server with high-speed interconnects (e.g., NVLink and NVSwitch). However, the rapidly increasing scale of graphs poses a challenge to applying GNN to real-world applications, due to limited GPU memory. This paper presents XGNN, a multi-GPU GNN training system that fully utilizes system memory (e.g., GPU and host memory), as well as high-speed interconnects. The core design of XGNN is the Global GNN Memory Store (GGMS), which abstracts underlying resources to provide a unified memory store for GNN training. It partitions hybrid input data, including graph topological and feature data, across both GPU and host memory. GGMS also provides easy-to-use APIs for GNN applications to access data transparently, forwarding data access requests to the actual physical data partitions automatically. Evaluation on various multi-GPU platforms using three common GNN models with four large-scale datasets shows that XGNN outperforms DGL, Quiver and DGL+C by up to 7.9X (from 2.3X), 15.7X (from 3.3X) and 2.8X (from 1.3X), respectively.
Zitieren
APA
Tang, D., Wang, J., Chen, R., Wang, L., Yu, W., Zhou, J., & Li, K. (2024). XGNN: Boosting Multi-GPU GNN Training via Global GNN Memory Store. Proceedings of the VLDB Endowment, 17(5), 1105-1118. https://doi.org/10.14778/3641204.3641219
Vancouver
Tang D, Wang J, Chen R, Wang L, Yu W, Zhou J, et al. XGNN: Boosting Multi-GPU GNN Training via Global GNN Memory Store. Proceedings of the VLDB Endowment. 2024;17(5):1105-1118. doi:10.14778/3641204.3641219
BibTeX
@article{Tang2024,
author = {Tang, Dahai and Wang, Jiali and Chen, Rong and Wang, Lei and Yu, Wenyuan and Zhou, Jingren and Li, Kenli},
title = {XGNN: Boosting Multi-GPU GNN Training via Global GNN Memory Store},
journal = {Proceedings of the VLDB Endowment},
year = {2024},
volume = {17},
number = {5},
pages = {1105--1118},
doi = {10.14778/3641204.3641219}
}
EUVIMED