Accelerating the Dutch Atmospheric Large-Eddy Simulation (DALES) model with OpenACC

TL;DR

This paper demonstrates how porting the Dutch Atmospheric Large-Eddy Simulation (DALES) model to GPUs using OpenACC directives improves performance, scalability, and portability across different hardware, with additional kernel optimization.

Contribution

The paper introduces a GPU acceleration approach for DALES using OpenACC, enabling high-performance atmospheric simulations on modern HPC hardware.

Findings

Significant reduction in simulation time on GPU-accelerated nodes

Effective weak scaling across multiple GPU hardware platforms

Enhanced kernel performance through auto-tuning with Kernel Tuner

Abstract

This paper presents the GPU porting through OpenACC directives of the Dutch Atmospheric Large-Eddy Simulation (DALES) application, a high-resolution atmospheric model. The code is written in Fortran~90 and features parallel (distributed) execution through spatial domain decomposition. We assess the performance of the GPU offloading, comparing the time-to-solution on regular and accelerated HPC nodes. %comparing the computational time between distributed and accelerated nodes. A weak scaling analysis is conducted and portability across NVIDIA A100 and H100 hardware %and AMD hardware is discussed. Finally, we show how targeted kernels can benefit from further optimization with Kernel Tuner, a GPU kernels auto-tuning package.