Developing Extensible Lattice-Boltzmann Simulators for General-Purpose Graphics-Processing Units
Stuart D. C. Walsh 1*, Martin O. Saar 21 Lawrence Livermore National Laboratory, Livermore, California, USA.
2 Department of Earth Sciences, University of Minnesota, Minneapolis, Minnesota, USA.
Received 31 October 2011; Accepted (in revised version) 26 January 2012
Available online 29 August 2012
Lattice-Boltzmann methods are versatile numerical modeling techniques capable of reproducing a wide variety of fluid-mechanical behavior. These methods are well suited to parallel implementation, particularly on the single-instruction multiple data (SIMD) parallel processing environments found in computer graphics processing units (GPUs). Although recent programming tools dramatically improve the ease with which GPU-based applications can be written, the programming environment still lacks the flexibility available to more traditional CPU programs. In particular, it may be difficult to develop modular and extensible programs that require variable on-device functionality with current GPU architectures. This paper describes a process of automatic code generation that overcomes these difficulties for lattice-Boltzmann simulations. It details the development of GPU-based modules for an extensible lattice-Boltzmann simulation package - LBHydra. The performance of the automatically generated code is compared to equivalent purpose written codes for both single-phase, multiphase, and multicomponent flows. The flexibility of the new method is demonstrated by simulating a rising, dissolving droplet moving through a porous medium with user generated lattice-Boltzmann models and subroutines.
Notice: Undefined variable: ams in /var/www/html/issue/abstract/readabs.php on line 163
PACS: 47.11.-j, 07.05.Bx
Key words: Lattice-Boltzmann methods, graphics processing units, computational fluid dynamics.
Email: firstname.lastname@example.org (S. D. C. Walsh), email@example.com (M. O. Saar)