2023/09/28 by Min, Misun, Lan, Yu-Hsiang, Fischer, Paul +2
#35-04 #D.0 #Distributed #F.2 #FOS: Computer and information sciences #G.2 #G.4 #I.6 #J.2 #Parallel #Performance (cs.PF) #and Cluster Computing (cs.DC)
paper · doi:10.48550/arxiv.2309.16381
We demonstrate NekRS performance results on various advanced GPU architectures. NekRS is a GPU-accelerated version of Nek5000 that targets high performance on exascale platforms. It is being developed in DOE's Center of Efficient Exascale Discretizations, which is one of the co-design centers under the Exascale Computing Project. In this paper, we consider Frontier, Crusher, Spock, Polaris, Perlmutter, ThetaGPU, and Summit. Simulations are performed with 17x17 rod-bundle geometries from small modular reactor applications. We discuss strong-scaling performance and analysis.