Large-Scale direct numerical simulations of turbulence using GPUs and modern Fortran-Reference-Cited by-同舟云学术

Large-Scale direct numerical simulations of turbulence using GPUs and modern Fortran

Published:2023-02-22 Issue:5 Volume:37 Page:487-502
ISSN:1094-3420
Container-title:The International Journal of High Performance Computing Applications
language:en
Short-container-title:The International Journal of High Performance Computing Applications

Author:

Karp Martin¹^ORCID,Massaro Daniele²^ORCID,Jansson Niclas³^ORCID,Hart Alistair⁴,Wahlgren Jacob¹,Schlatter Philipp²^ORCID,Markidis Stefano¹

Affiliation:

1. Division of Computational Science and Technology, EECS, KTH Royal Institute of Technology, Stockholm, Sweden

2. SimEx/FLOW, Engineering Mechanics, KTH Royal Institute of Technology, Stockholm, Sweden

3. PDC Centre for High Performance Computing, EECS, KTH Royal Institute of Technology, Stockholm, Sweden

4. Hewlett Packard Enterpise (HPE), UK

Abstract

We present our approach to making direct numerical simulations of turbulence with applications in sustainable shipping. We use modern Fortran and the spectral element method to leverage and scale on supercomputers powered by the Nvidia A100 and the recent AMD Instinct MI250X GPUs, while still providing support for user software developed in Fortran. We demonstrate the efficiency of our approach by performing the world’s first direct numerical simulation of the flow around a Flettner rotor at Re = 30,000 and its interaction with a turbulent boundary layer. We present a performance comparison between the AMD Instinct MI250X and Nvidia A100 GPUs for scalable computational fluid dynamics. Our results show that one MI250X offers performance on par with two A100 GPUs and has a similar power efficiency based on readings from on-chip energy sensors.

Funder

Vetenskapsrådet

Swedish e-Science Research Centre

Publisher

SAGE Publications

Subject

Hardware and Architecture,Theoretical Computer Science,Software

Link

http://journals.sagepub.com/doi/pdf/10.1177/10943420231158616

Reference51 articles.

1. MFEM: A modular finite element methods library

2. The deal.II library, Version 9.2

3. Uniform blowing and suction applied to nonuniform adverse-pressure-gradient wing boundary layers

4. Scientific multi-agent reinforcement learning for wall-models of turbulent flows

5. Experiments on a Flettner rotor at critical and supercritical Reynolds numbers

Cited by 5 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. A comprehensive framework to enhance numerical simulations in the spectral-element code Nek5000;Computer Physics Communications;2024-09

2. Direct numerical simulation of the turbulent flow around a Flettner rotor;Scientific Reports;2024-02-06

3. Uncertainty Quantification of Reduced-Precision Time Series in Turbulent Channel Flow;Proceedings of the SC '23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis;2023-11-12

4. Exploring the Ultimate Regime of Turbulent Rayleigh–Bénard Convection Through Unprecedented Spectral-Element Simulations;Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis;2023-11-11

5. Corrigendum to large-scale direct numerical simulations of turbulence using GPUs and modern Fortran;The International Journal of High Performance Computing Applications;2023-05-05