A Flexible Patch-Based Lattice Boltzmann Parallelization Approach for Heterogeneous GPU-CPU Clusters
Preprint
Abstract
Sustaining a large fraction of single GPU performance in parallel computations is considered to be the major problem of GPU-based clusters. In this article, this topic is addressed in the context of a lattice Boltzmann flow solver that is integrated in the WaLBerla software framework. We propose a multi-GPU implementation using a block-structured MPI parallelization, suitable for load balancing and heterogeneous computations on CPUs and GPUs. The overhead required for multi-GPU simulations is discussed in detail and it is demonstrated that the kernel performance can be sustained to a large extent. With our GPU implementation, we achieve nearly perfect weak scalability on InfiniBand clusters. However, in strong scaling scenarios multi-GPUs make less efficient use of the hardware than IBM BG/P and x86 clusters. Hence, a cost analysis must determine the best course of action for a particular simulation task. Additionally, weak scaling results of heterogeneous simulations conducted on CPUs and GPUs simultaneously are presented using clusters equipped with varying node configurations.Keywords
All Related Versions
- Version 1, 2010-07-08, ArXiv
- Published version: Parallel Computing, 37 (9), 536.
This publication has 10 references indexed in Scilit:
- A new approach to the lattice Boltzmann method for graphics processing unitsComputers & Mathematics with Applications, 2011
- Coupling multibody dynamics and computational fluid dynamics on 8192 processor coresParallel Computing, 2010
- Lattice-Boltzmann Method for Complex FlowsAnnual Review of Fluid Mechanics, 2010
- BENCHMARK ANALYSIS AND APPLICATION RESULTS FOR LATTICE BOLTZMANN SIMULATIONS ON NEC SX VECTOR AND INTEL NEHALEM SYSTEMSParallel Processing Letters, 2009
- A flexible high‐performance Lattice Boltzmann GPU code for the simulations of fluid flows in complex geometriesConcurrency and Computation: Practice and Experience, 2009
- TeraFLOP computing on a desktop PC with GPUs for 3D CFDInternational Journal of Computational Fluid Dynamics, 2008
- Statistical mechanics of the fluctuating lattice Boltzmann equationPhysical Review E, 2007
- On the single processor performance of simple lattice Boltzmann kernelsComputers & Fluids, 2006
- Lattice Boltzmann Model for the Incompressible Navier–Stokes EquationJournal of Statistical Physics, 1997
- A spiral model of software development and enhancementACM SIGSOFT Software Engineering Notes, 1986