2022/11/28 by Omlin, Samuel, Räss, Ludovic, Utkin, Ivan · 1 citation
#Computational Physics (physics.comp-ph) #Distributed #FOS: Computer and information sciences #FOS: Physical sciences #Geophysics (physics.geo-ph) #Parallel #and Cluster Computing (cs.DC)
paper · doi:10.48550/arxiv.2211.15716
We present a straightforward approach for distributed parallelization of stencil-based xPU applications on a regular staggered grid, which is instantiated in the package ImplicitGlobalGrid.jl. The approach allows to leverage remote direct memory access and enables close to ideal weak scaling of real-world applications on thousands of GPUs. The communication costs can be easily hidden behind computation.