An unstructured CFD mini-application for the performance prediction of a production CFD code

Research output: Contribution to journalArticlepeer-review

Authors

  • A. M.B. Owenson
  • S. A. Wright
  • R. A. Bunt
  • Y. K. Ho
  • M. J. Street

Colleges, School and Institutes

External organisations

  • University of Warwick
  • University of York
  • Rolls Royce PLC

Abstract

Maintaining the performance of large scientific codes is a difficult task. To aid in this task, a number of mini-applications have been developed that are more tractable to analyze than large-scale production codes while retaining the performance characteristics of them. These “mini-apps” also enable faster hardware evaluation and, for sensitive commercial codes, allow evaluation of code and system changes outside of access approval processes. In this paper, we develop MG-CFD, a mini-application that represents a geometric multigrid, unstructured computational fluid dynamics (CFD) code, designed to exhibit similar performance characteristics without sharing commercially sensitive code. We detail our experiences of developing this application using guidelines detailed in existing research and contributing further to these. Our application is validated against the inviscid flux routine of HYDRA, a CFD code developed by Rolls-Royce plc for turbomachinery design. This paper (1) documents the development of MG-CFD, (2) introduces an associated performance model with which it is possible to assess the performance of HYDRA on new HPC architectures, and (3) demonstrates that it is possible to use MG-CFD and the performance models to predict the performance of HYDRA with a mean error of 9.2% for strong-scaling studies.

Bibliographic note

Funding Information: This research is supported by Rolls-Royce plc, by the EU Horizon 2020 Clean Sky Project, by the UK Engineering and Physical Sciences Research Council (EPSRC), and by the Intel Corporation: (EP/S005072/1 - Strategic Partnership in Computational Science for Advanced Simulation and Modelling of Engineering Systems - ASiMoV; EPSRC Industrial CASE award 15220082). The authors would like to thank Rolls-Royce plc for granting permission to publish this work. Publisher Copyright: © 2019 The Authors Concurrency and Computation: Practice and Experience Published by John Wiley & Sons Ltd.

Details

Original languageEnglish
Article numbere5443
JournalConcurrency Computation
Volume32
Issue number10
Publication statusPublished - 25 May 2020

Keywords

  • computational fluid dynamics, high performance computing, mini-application, performance analysis, performance modeling, scientific computing