MQ-GNN通过多队列流水线加速图神经网络训练，提升多卡效率。
MQ-GNN: A Multi-Queue Pipelined Architecture for Scalable and Efficient GNN Training
- 采用多队列流水线和异步模型更新机制，重叠训练各阶段
- 实测比基线快4.6倍，GPU利用率提升30%且精度不降
- 适合大规模图神经网络的多机多卡高效训练场景
图神经网络（GNN）在学习图结构数据方面表现强大，但其可扩展性受限于低效的小批量生成、数据传输瓶颈以及昂贵的跨GPU同步。现有训练框架无法重叠这些阶段，导致资源利用率低下。本文提出MQ-GNN，一种多队列流水线架构，通过交错执行GNN训练阶段并优化资源利用来最大化训练效率。MQ-GNN引入了就绪更新异步一致模型（RaCoM），实现异步梯度共享与模型更新，同时通过自适应周期性同步保证全局一致性。此外，采用全局邻居采样与缓存机制降低数据传输开销，并设计自适应队列大小策略以平衡计算与内存效率。在四个大规模数据集和十种基线模型上的实验表明，MQ-GNN实现了最高4.6倍的训练速度提升，GPU利用率提高30%，且保持了具有竞争力的准确率。结果证明MQ-GNN是多GPU GNN训练中可扩展且高效的解决方案。
原文摘要 · Abstract (English)
Graph Neural Networks (GNNs) are powerful tools for learning graph-structured data, but their scalability is hindered by inefficient mini-batch generation, data transfer bottlenecks, and costly inter-GPU synchronization. Existing training frameworks fail to overlap these stages, leading to suboptimal resource utilization. This paper proposes MQ-GNN, a multi-queue pipelined framework that maximizes training efficiency by interleaving GNN training stages and optimizing resource utilization. MQ-GNN introduces Ready-to-Update Asynchronous Consistent Model (RaCoM), which enables asynchronous gradient sharing and model updates while ensuring global consistency through adaptive periodic synchronization. Additionally, it employs global neighbor sampling with caching to reduce data transfer overhead and an adaptive queue-sizing strategy to balance computation and memory efficiency. Experiments on four large-scale datasets and ten baseline models demonstrate that MQ-GNN achieves up to \boldmath $\bm{4.6\,\times}$ faster training time and 30% improved GPU utilization while maintaining competitive accuracy. These results establish MQ-GNN as a scalable and efficient solution for multi-GPU GNN training.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。