arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2410.01657cs.DCcs.LGphysics.comp-ph
收藏

面向分布式基于网格的数据驱动建模的可扩展且一致的图神经网络

Scalable and Consistent Graph Neural Networks for Distributed Mesh-based Data-driven Modeling

Argonne National Laboratory(阿贡国家实验室);Pennsylvania State University(宾夕法尼亚州立大学)

Shivam Barwey, Riccardo Balin, Bethany Lusch, Saumil Patel, Ramesh Balakrishnan, Pinaki Pal, Romit Maulik, Venkatram Vishwanath

AI总结:

提出采用一致神经消息传递层和边界晕节点的分布式GNN方法,并结合NekRS实现可扩展的基于网格的数据驱动建模,在Frontier上扩展至O(1B)节点。

摘要:

本工作开发了一种用于基于网格的建模应用的分布式图神经网络(GNN)方法,该方法采用一致的神经消息传递层。顾名思义,其重点是通过子图边界处的晕节点(halo nodes)实现满足物理一致性的可扩展操作。这里的一致性是指,在单个秩(单个大图)上训练和评估的GNN,与在多个秩(分区图)上的评估在算术上等价。通过将GNN与NekRS(由阿贡国家实验室开发的支持GPU的百亿亿次计算流体力学(CFD)求解器)相接口,这一概念得到了演示。文中展示了如何将NekRS网格划分与分布式GNN训练和推理例程相连接,从而形成可扩展的基于网格的数据驱动建模工作流。我们研究了一致性对基于网格的GNN可扩展性的影响,在Frontier百亿亿次超级计算机上证明了一致GNN在多达O(1B)个图节点时的高效扩展。

英文摘要:

This work develops a distributed graph neural network (GNN) methodology for mesh-based modeling applications using a consistent neural message passing layer. As the name implies, the focus is on enabling scalable operations that satisfy physical consistency via halo nodes at sub-graph boundaries. Here, consistency refers to the fact that a GNN trained and evaluated on one rank (one large graph) is arithmetically equivalent to evaluations on multiple ranks (a partitioned graph). This concept is demonstrated by interfacing GNNs with NekRS, a GPU-capable exascale CFD solver developed at Argonne National Laboratory. It is shown how the NekRS mesh partitioning can be linked to the distributed GNN training and inference routines, resulting in a scalable mesh-based data-driven modeling workflow. We study the impact of consistency on the scalability of mesh-based GNNs, demonstrating efficient scaling in consistent GNNs for up to O(1B) graph nodes on the Frontier exascale supercomputer.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑