水利学报2025，Vol.56Issue(12)：1647-1658,12.DOI:10.13243/j.cnki.slxb.20250262
二维水动力模型多GPU分布式数据并行计算方法研究
Research on multi-GPU distributed data parallel computing methods for 2D hydrodynamic models
摘要
Abstract
To address the computational bottlenecks of 2D hydrodynamic models for flood simulation of complex watersheds,this study develops a multi-GPU heterogeneous parallel architecture based on physical topology-based domain partitioning and integration of distributed data parallelism with NCCL asynchronous communication,enabling supercomputing-level acceleration for unstructured triangular mesh-based hydrodynamic models.Through a physical topology-preserving domain partitioning algorithm,dynamic load balancing across multiple GPUs was achieved,while ensuring the integrity of adjacency relationships within unstructured triangular meshes.By combining a distrib-uted data parallel solution framework based on the Godunov finite volume method with an NCCL asynchronous com-munication strategy,a heterogeneous acceleration architecture with"computation-communication"collaboration was established.Experimental validation shows that,in a 2D idealized dam-break test case,the model accurately cap-tured the propagation characteristics of the dam-break shock waves,maintaining high consistency with theoretical solutions.In the case study of the Laoshan Reservoir dam-break flood simulation,the 2-hour flood evolution simula-tion was completed in just 15.70 s using 8 GPUs for parallel computing,achieving a speedup of 308.43 times.Fur-thermore,the data communication efficiency between subdomains improved by 8.67%compared to traditional MPI-based methods.The proposed architecture supports elastic scaling from single-node to cross-node GPU clusters,pro-viding a supercomputing-level hydrodynamic engine with second-level response capabilities for digital twin water-sheds.It offers core technological support for advancing flood forecasting and dispatching from static planning toward dynamic rehearsal paradigms.关键词
二维水动力模型/有限体积法/计算域分割/NCCL/异构计算Key words
2D hydrodynamic model/finite volume method/computational domain partitioning/NCCL communica-tion/heterogeneous parallel computing分类
建筑与水利引用本文复制引用
丁武,杨芳,王卫光,王汉岗,何用,蔺崇哲..二维水动力模型多GPU分布式数据并行计算方法研究[J].水利学报,2025,56(12):1647-1658,12.基金项目
国家重点研发计划项目(2024YFC3212000) （2024YFC3212000）
