Abstract
Recently, several research groups have demonstrated significant speedups of scientific computations using General Purpose Graphics Processor Units (GPGPU) as massively-parallel “co-processors” to the Central Processing Unit (CPU). However, the tremendous computational power of GPGPUs has come with a high price since their implementation to Computational Fluids Dynamics (CFD) solvers is still a challenge. To achieve this implementation, the RapidCFD library was developed from the Open Field Operation and Manipulation (OpenFOAM) CFD software to let that the multi-GPGPU were able of running almost the entire simulation in parallel. The parallel performance, as fixed-size speed-up, efficiency and parallel fraction, according to the Amdahl’s law, were compared in two massively parallel multi-GPGPU architectures using Nvidia Tesla C1060 and M2090 units. The simulations were executed on a 3D turbo-machinery benchmark which consist of a structured grid domain of 1 million cells. The results obtained from the implementation of the new library on different software and hardware layouts show that by transferring directly all the computations executed by the linear system solvers to the GPGPU, is possible to make a typical CFD simulation until 9 times faster. Additionally a grid convergence analysis and pressure recovery measurements were executed over scaled computational domains. Thus, it is expected to obtain an affordable low computational cost when the domain be scaled in order to achieve a high flow resolution.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Niemeyer, K.E., Sung, C.-J.: Recent progress and challenges in exploiting graphics processors in computational fluid dynamics. J. Supercomput. 67(2), 528–564 (2014)
Navarro, C., Hitschfeld-Kahler, N., Mateu, L.: A survey on parallel computing and its applications in data-parallel problems using GPU architectures. Commun. Comput. Phys. 15(2), 285–329 (2014)
Posey, S., See, S., Wang, M.: GPU progress and directions in applied CFD. In: Eleventh International Conference on CFD in the Minerals and Process Industries, Melbourne, Australia (2015)
AlOnazi, A.: Design and optimization of OpenFOAM-based CFD applications for modern hybrid and heterogeneous HPC platforms. Master thesis, King Abdullah University of Science and Technology, Thuwal, Kingdom of Saudi Arabia (2014)
NVIDIA Corporation, Cuda C Programming Guide v6.5 (2014)
Aissa, M.: GPU-accelerated CFD simulations for turbomachinery design optimization. Doctoral thesis, Delft University of Technology (2017)
Gebart, B., Gustavsson, L., Karlsson, R.: Proceedings of Turbine 99 Workshop on Draft Tube Flow in Porjus, Sweden, Luleä University of Technology (2000)
Engström, T., Gustavsson, L., Karlsson, R.: Turbine-99 workshop 2 on draft tube flow. In: Proceedings of 21st IAHR Symposium on Hydraulic Machinery and Systems, Lausanne, Switzerland (2005)
Cervantes, M., Engstöm, T., Gustavsson, L.: Proceedings of the Third IAHR/ERCOFTAC Workshop on Draft Tube Flows, Luleå University of Technology, Porjus, Sweden (2005)
Galván, S., Reggio, M., Guibault, F.: Assessment study of k-ε turbulence models and near-wall modeling for steady state swirling flow analysis in draft tube using fluent. Eng. Appl. Comput. Fluid Mech. 5(4), 459–478 (2011)
Jasiński, D.: Adapting OpenFOAM for massively parallel GPU architecture. In: The 3rd OpenFOAM User Conference, Stuttgart, Germany (2015)
simFlow CFD software, Atizar/RapidCFD-dev, GitHub, Inc. https://github.com/Atizar/RapidCFD-dev
Afzal, A., Ansari, Z., Faizabadi, A.R., Ramis, M.K.: Parallelization strategies for computational fluid dynamics software: state of the art review. Arch. Comput. Methods Eng. 24(2), 337–363 (2017)
OpenCFD Limited, OpenFOAM. The Open Source CFD Tool Box. User Guide v1706 (2017)
Khajeh-Saeed, A., Perot, J.B.: Computational fluid dynamics simulations using many graphics processors. Comput. Sci. Eng. 14, 10–19 (2012)
He, Q., Hongli, C., Jingchao, F.: Acceleration of the OpenFOAM-based MHD solver using graphics processing units. Fusion Eng. Des. 101, 88–93 (2015)
Malecha, Z., et al.: GPU-based simulation of 3D blood flow in abdominal aorta using OpenFOAM. Arch. Mech. 63(2), 137–161 (2011)
Pacheco, P.: An Introduction to Parallel Programming. Elsevier, Amsterdam (2011)
McCool, M., Robison, A.D., Reinders, J.: Structured Parallel Programming: Patterns for Efficient Computation. Morgan Kaufmann Publishers, Waltham (2012)
Andersson, U.: Test case T- some news results and updates since workshop 1. In: Proceedings of Turbine 99-WS2, the Second ERCOFTAC Workshop on Draft Tubeflow, Alvkarleby, Sweden (2001)
Herrera, N., Galván, S., Camacho, J., Solorio, G., Aguilar, A.: Automatic shape optimization of a conical duct diffuser using a distributed computing algorithm. J. Braz. Soc. Mech. Sci. Eng. 39(11), 4367–4378 (2017)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Molinero, D., Galván, S., Pacheco, J., Herrera, N. (2019). Multi GPU Implementation to Accelerate the CFD Simulation of a 3D Turbo-Machinery Benchmark Using the RapidCFD Library. In: Torres, M., Klapp, J. (eds) Supercomputing. ISUM 2019. Communications in Computer and Information Science, vol 1151. Springer, Cham. https://doi.org/10.1007/978-3-030-38043-4_15
Download citation
DOI: https://doi.org/10.1007/978-3-030-38043-4_15
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-38042-7
Online ISBN: 978-3-030-38043-4
eBook Packages: Computer ScienceComputer Science (R0)