WaLBerla: Exploiting Massively Parallel Systems for Lattice Boltzmann Simulations
In this chapter, a software concept for massively parallel computational fluid dynamics (CFD) applications is introduced. The focus thereby lies on the parallelization, which is based on a domain partitioning scheme named patch concept. This concept also enables a seamless specialization of the partitions to different application features as well as the possibility for further optimization such as memory reduction. It is discussed in detail how our design ensures an efficient and flexible implementation. The suitability and efficiency of this concept is demonstrated and evaluated with the waLBerla project, which aims at the development of an efficient massively parallel lattice Boltzmann framework providing the necessary features for several CFD applications. To discuss the suitability of the parallelization for massively parallel usage, various test scenarios have been investigated on different architectures. These tests include serial, weak and strong scaling experiments up to 810 cores and up to a domain size of 15303 lattice cells.
KeywordsMessage Passing Interface Lattice Boltzmann Method Lattice Boltzmann Model Lattice Boltzmann Equation Lattice Boltzmann Simulation
Unable to display preview. Download preview PDF.
- 1.C. Feichtinger, J. Götz, S. Donath, K. Iglberger, U. Rüde, Concepts of waLBerla prototype 0.1, Tech. Rep. 07–10, University of Erlangen-Nuremberg, Computer Science 10 – Systemsimulation (2007).Google Scholar
- 5.M. Stürmer, J. Götz, G. Richter, A. Dörfler, U. Rüde, Blood flow simulation on the Cell Broadband Engine using the lattice Boltzmann method, Tech. Rep. 07–9, University of Erlangen- Nuremberg, Computer Science 10 – Systemsimulation, submitted to the International Conference for Mesoscopic Methods in Engineering and Science, ICMMES (2007).Google Scholar
- 8.Information on fuel cells, http://www.fuelcells.org (2008).
- 9.Information on the HLRB II, http://www.lrz-muenchen.de/services/compute/hlrb/ (2008).
- 14.T. Zeiser, H.-J. Freund, J. Bernsdorf, P. Lammers, G. Brenner, F. Durst, Detailed Simulation of Transport Processes in Reacting Multi-Species Flows Through Complex Geometries by Means of the Lattice Boltzmann method, in: In High Performance Computing in Science and Engineering ’01, Transactions of the High Performance Computing Center Stuttgart (HLRS), Springer (2002).Google Scholar
- 17.C. Körner, T. Pohl, U. Rüde, N. Thürey, T. Hofmann, FreeWIHR: Lattice Boltzmann methods with free surfaces and their application in material technology, in: A. Bode, F. Durst (Eds.), High Performance Computing in Science and Engineering, Garching 2004, Springer (2005), pp. 225–236.Google Scholar
- 18.C. Körner, T. Pohl, U. Rüde, N. Thürey, T. Zeiser, Parallel Lattice Boltzmann Methods for CFD Applications, in: A. Bruaset, A. Tveito (Eds.), Numerical Solution of Partial Differential Equations on Parallel Computers, Vol. 51 of Lecture Notes for Computational Science and Engineering, Springer (2005) Ch. 5, pp. 439–465.Google Scholar
- 19.J. Wilke, T. Pohl, M. Kowarschik, U. Rüde, Cache Performance Optimizations for Parallel Lattice Boltzmann Codes, in: Proc. of the EuroPar-03 Conf., Vol. 2790 of Lecture Notes in Computer Science, Springer (2003), pp. 441–450.Google Scholar
- 21.D. Hänel, Molekulare Gasdynamik, Springer (2004).Google Scholar
- 22.D. Wolf-Gladrow, Lattice-Gas Cellular Automata and Lattice Boltzmann Models, Springer (2000).Google Scholar
- 25.J. Götz, Numerical Simulation of Blood Flow with Lattice Boltzmann Methods, Master’s thesis, University of Erlangen-Nuremberg, Computer Science 10 – Systemsimulation (2006).Google Scholar
- 26.Information on the Juelicher Initiative Cell Cluster (JUICE), http://www.fz-juelich.de/jsc/service/juice (2008).
- 27.Top500, The top 500 supercomputer sites, http://www.top500.org (2008).
- 28.T. Zeiser, J. Götz, M. Stürmer, On performance and accuracy of lattice Boltzmann approaches for single phase flow in porous media: A toy became an accepted tool – How to maintain its features despite more and more complex (physical) models and changing trends in high performance computing!?On performance and accuracy of lattice Boltzmann approaches for single phase flow in porous media, in: Proceedings of 3rd Russian-GermanWorkshop on High Performance Computing, Novosibirsk, Springer (2008).Google Scholar
- 29.J. D. McCalpin, STREAM: Sustainable memory bandwidth in high performance computers, http://www.cs.virginia.edu/stream/ (1991–2008).
- 30.T. Zeiser, Private correspondence with Thomas Zeiser, Regional Computing Center Erlangen (RRZE) (Aug. 2008).Google Scholar