一句话：提出 Betty，用批级图划分、冗余嵌入图划分和内存感知划分缓解 GNN 训练显存瓶颈，支持单 GPU 大规模训练。
图神经网络（GNN）可通过更先进的聚合器、更深的聚合深度和更大的采样率提升效果，但这会显著增加显存占用，常超出 GPU 容量。与 DNN 训练不同，GNN 的 mini-batch 中输入特征与输出标签之间存在复杂依赖，使批级划分困难。为此，论文提出 Betty，一种基于批级图划分的可扩展 GNN 训练方法，并引入冗余嵌入图（REG）划分与内存感知划分两项技术，以缓解分区间的冗余和负载不均衡。作者在大规模真实数据集上评估，结果表明 Betty 能显著缓解显存瓶颈，在单 GPU 上支持更深聚合、更高采样率、更大训练 batch 及更先进聚合器的 GNN 训练。
The Graph Neural Network (GNN) is showing outstanding results in improving the performance of graph-based applications. Recent studies demonstrate that GNN performance can be boosted via using more advanced aggregators, deeper aggregation depth, larger sampling rate, etc. While leading to promising results, the improvements come at a cost of significantly increased memory footprint, easily exceeding GPU memory capacity. In this paper, we introduce a method, Betty, to make GNN training more scalable and accessible via batch-level partitioning. Different from DNN training, a mini-batch in GNN has complex dependencies between input features and output labels, making batch-level partitioning difficult. Betty introduces two noveltechniques, redundancy-embedded graph (REG) partitioning and memory-aware partitioning, to effectively mitigate the redundancy and load imbalances issues across the partitions. Our evaluation of large-scale real-world datasets shows that Betty can significantly mitigate the memory bottleneck, enabling scalable GNN training with much deeper aggregation depths, larger sampling rate, larger training batch sizes, together with more advanced aggregators, with a few as a single GPU.