GPU-HADVPPM4HIP V1.0: using the heterogeneous-compute interface for portability (HIP) to speed up the piecewise parabolic method in the CAMx (v6.10) air quality model on China's domestic GPU-like accelerator

Abstract

Graphics processing units (GPUs) are becoming a compelling acceleration strategy for geoscience numerical models due to their powerful computing performance. In this study, AMD's heterogeneous-compute interface for portability (HIP) was implemented to port the GPU acceleration version of the piecewise parabolic method (PPM) solver (GPU-HADVPPM) from NVIDIA GPUs to China's domestic GPU-like accelerators like GPU-HADVPPM4HIP. Further, it introduced the multi-level hybrid parallelism scheme to improve the total computational performance of the HIP version of the CAMx (Comprehensive Air Quality Model with Extensions; CAMx-HIP) model on China's domestic heterogeneous cluster. The experimental results show that the acceleration effect of GPU-HADVPPM on the different GPU accelerators is more apparent when the computing scale is more extensive, and the maximum speedup of GPU-HADVPPM on the domestic GPU-like accelerator is 28.9×faster. The hybrid parallelism with a message passing interface (MPI) and HIP enables achieving up to a 17.2× speedup when configuring 32 CPU cores and GPU-like accelerators on the domestic heterogeneous cluster. The OpenMP technology is introduced further to reduce the computation time of the CAMx-HIP model by 1.9×. More importantly, by comparing the simulation results of GPU-HADVPPM on NVIDIA GPUs and domestic GPU-like accelerators, it is found that the simulation results of GPU-HADVPPM on domestic GPU-like accelerators have less difference than the NVIDIA GPUs. Furthermore, we also show that the data transfer efficiency between CPU and GPU has a meaningful essential impact on heterogeneous computing and point out that optimizing the data transfer efficiency between CPU and GPU is one of the critical directions to improve the computing efficiency of geoscience numerical models in heterogeneous clusters in the future.