EUVIMEDEUVIMED entdecken
Crossref

OUTRE: An OUT-of-Core De-REdundancy GNN Training Framework for Massive Graphs within A Single Machine

Zeang Sheng, Wentao Zhang, Yangyu Tao, Bin Cui

Proceedings of the VLDB Endowment · 2024-07 · 17 (11) · 2960-2973

DOI: 10.14778/3681954.3681976 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the VLDB Endowment, online erschienen am 30.08.2024. Verfasst von Zeang Sheng und 3 weiteren Autor:innen. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 61 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Sampling-based Graph Neural Networks (GNNs) have become the de facto standard for handling various graph learning tasks on large-scale graphs. As the graph size grows larger and even exceeds the standard host memory size of a single machine, out-of-core sampling-based GNN training has gained attention from the community. For out-of-core sampling-based GNN training, the performance bottleneck is the data preparation process that includes sampling neighbor lists and gathering node features from external storage. Based on this observation, existing out-of-core GNN training frameworks try to accomplish larger percentages of data requests without inquiring the external storage by designing better in-memory caches. However, the enormous overall requested data volume is unchanged under this approach. In this paper, we present a new perspective on reducing the overall requested data volume. Through a quantitative analysis, we find that Neighborhood Redundancy and Temporal Redundancy exist in out-of-core sampling-based GNN training. To reduce these two kinds of data redundancies, we propose OUTRE, an OUT-of-core de-REdundancy GNN training framework. OUTRE incorporates two new designs, partition-based batch construction and historical embedding cache , to reduce the corresponding data redundancies. Moreover, we propose automatic cache space management to automatically organize available memory for different caches. Evaluation results on four public large-scale graph datasets show that OUTRE achieves 1.52× to 3.51× speedup against the SOTA framework.

Zitieren

APA

Sheng, Z., Zhang, W., Tao, Y., & Cui, B. (2024). OUTRE: An OUT-of-Core De-REdundancy GNN Training Framework for Massive Graphs within A Single Machine. Proceedings of the VLDB Endowment, 17(11), 2960-2973. https://doi.org/10.14778/3681954.3681976

Vancouver

Sheng Z, Zhang W, Tao Y, Cui B. OUTRE: An OUT-of-Core De-REdundancy GNN Training Framework for Massive Graphs within A Single Machine. Proceedings of the VLDB Endowment. 2024;17(11):2960-2973. doi:10.14778/3681954.3681976

BibTeX
@article{Sheng2024,
  author = {Sheng, Zeang and Zhang, Wentao and Tao, Yangyu and Cui, Bin},
  title = {OUTRE: An OUT-of-Core De-REdundancy GNN Training Framework for Massive Graphs within A Single Machine},
  journal = {Proceedings of the VLDB Endowment},
  year = {2024},
  volume = {17},
  number = {11},
  pages = {2960--2973},
  doi = {10.14778/3681954.3681976}
}
Kostenlos starten →