On floating point precision in computational fluid dynamics using OpenFOAM

TL;DR

This paper investigates how reducing floating point precision affects performance and accuracy in computational fluid dynamics simulations using OpenFOAM, highlighting potential speedups and the importance of precision choice.

Contribution

It provides a comprehensive analysis of reduced precision impacts on CFD simulations in OpenFOAM, including a theoretical model for scalability and practical GPU implementation insights.

Findings

Reduced precision can significantly speed up CFD simulations.

Accuracy and convergence depend on the type of fluid and simulation complexity.

Scalability behavior varies with floating point precision and hardware specifics.

Abstract

Thanks to the computational power of modern cluster machines, numerical simulations can provide, with an unprecedented level of details, new insights into fluid mechanics. However, taking full advantage of this hardware remains challenging since data communication remains a significant bottleneck to reaching peak performances. Reducing floating point precision is a simple and effective way to reduce data movement and improve the computational speed of most applications. Nevertheless, special care needs to be taken to ensure the quality and convergence of computed solutions, especially when dealing with complex fluid simulations. In this work, we analyse the impact of reduced (single and mixed compared to double) precision on computational performance and accuracy for computational fluid dynamics. Using the open source library OpenFOAM, we consider incompressible, compressible, and multiphase fluid solvers for testing on relevant benchmarks for flows in the laminar and turbulent regime and in the presence of shock waves. Computational gain and changes in the scalability of applications in reduced precision are also discussed. In particular, an ad hoc theoretical model for the strong scaling allows us to interpret and understand the observed behaviors, as a function of floating point precision and hardware specifics. Finally, we show how reduced precision can significantly speed up a hybrid CPU-GPU implementation, made available to OpenFOAM end-users recently, that simply relies on a GPU linear algebra solver developed by hardware vendors.