Accelerated lattice Boltzmann simulation using GPU and OpenACC with data management. (June 2017)

• Record Type: Journal Article
• Authors: Xu, A.; Shi, L.; Zhao, T.S.
• Is Part Of: International journal of heat and mass transfer. Volume 109(2017)
• Page Start: 577
• Page End: 588
• Publication Date: 2017-06
• Subjects: GPU computing -- OpenACC -- Lattice Boltzmann method -- Heat and mass trasnfer; Heat -- Transmission -- Periodicals; Mass transfer -- Periodicals; Chaleur -- Transmission -- Périodiques; Transfert de masse -- Périodiques; Electronic journals; 621.4022
• DOI: 10.1016/j.ijheatmasstransfer.2017.02.032
• Languages: English
• ISSNs: 0017-9310

Abstract

Highlights: Improved implementation of OpenACC is adopted to accelerate LB simulation. A speedup around 50–60X can be achieved for multiphysics LB simulation. Benchmark quality results are obtained with fine grid of 2049 2 . Abstract: We assess the performance of the combined Open Accelerator (OpenACC) programming standard and graphics processing unit (GPU) accelerator for lattice Boltzmann (LB) simulations of fluid flow, heat and mass transfer. By optimizing the data layout, minimizing the memory access frequency, and adjusting the number of gangs and vector length, we show that the enhanced parallel computations can result in orders of magnitudes of speedup relative to the serial implementation of the LB algorithm. Based on such implementations, benchmark quality results are obtained with fine grid of 2049 2 for both two-dimensional lid driven cavity flow with Reynolds number up to 7500, and double diffusive cavity flow with solute Rayleigh number up to 10 8 .