Lattice-Boltzmann Simulation of the Shallow-Water Equations with Fluid-Structure Interaction on Multi- and Manycore Processors

Geveler, Markus; Ribbrock, Dirk; Göddeke, Dominik; Turek, Stefan

doi:10.1007/978-3-642-16233-6_11

Cited by 20 publications

(24 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The benchmarks of the solver configurations include comparisons with basic linear algebra kernels to evaluate different performance and memory characteristics. We refer to previously published work for an analysis in terms of mass conservation, a comparison with external codes and some other aspects not covered here [3,5].…”

Section: Resultsmentioning

confidence: 99%

“…We have evaluated our approach with a multigrid solver for the prototypical Poisson problem, and an explicit finite difference solver for the shallow water equations (SWE). In this paper, we go significantly further and apply newly developed and improved backends [3,4] to the more challenging task of interactive fluid dynamics (CFD) using a coupled Lattice-Boltzmann (LBM) fluid structure interaction (FSI) application. We aim at exploiting the full performance of hybrid single nodes, equipped with CPUs and GPUs or the Cell BE, an important prerequisite towards efficient large-scale computations.…”

Section: Paper Contribution and Paper Overviewmentioning

confidence: 99%

“…In comparison to previous publications, a new multi-core backend has been implemented from scratch, the GPU backend now supports multiple devices as well as transfers between them, and a new MPI backend [3,5] allows for the use of distributed memory systems. At the top level all backends targeting parallelism follow a similar design.…”

Section: Hardware Backendsmentioning

confidence: 99%

See 2 more Smart Citations

A simulation suite for Lattice-Boltzmann based real-time CFD applications exploiting multi-level parallelism on modern multi- and many-core architectures

Geveler

Ribbrock

Mallach

et al. 2011

Journal of Computational Science

Self Cite

View full text Add to dashboard Cite

We present a software approach to hardware-oriented numerics which builds upon an augmented, previously published set of open-source libraries facilitating portable code development and optimisation on a wide range of modern computer architectures. In order to maximise efficiency, we exploit all levels of parallelism, including vectorisation within CPU cores, the Cell BE and GPUs, shared memory thread-level parallelism between cores, and parallelism between heterogeneous distributed memory resources in clusters. To evaluate and validate our approach, we implement a collection of modular building blocks for the easy and fast assembly and development of CFD applications based on the shallow water equations: We combine the Lattice-Boltzmann method with fluid-structure interaction techniques in order to achieve real-time simulations targeting interactive virtual environments. Our results demonstrate that recent multi-core CPUs outperform the Cell BE, while GPUs are significantly faster than conventional multi-threaded SSE code. In addition, we verify good scalability properties of our application on small clusters.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Paper Contribution and Paper Overviewmentioning

confidence: 99%

See 1 more Smart Citation

A simulation suite for Lattice-Boltzmann based real-time CFD applications exploiting multi-level parallelism on modern multi- and many-core architectures

Geveler

Ribbrock

Mallach

et al. 2011

Journal of Computational Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…There are already GPU implementations as [17], where it was adapted using textures; but more recently, with the advent of general programmability of GPUs we find CUDA implementations like [18] and [19]. [20] proposed an alternative kernel implementation to reduce memory usage and [21] targeted multiple different parallel architectures using higher-level libraries for solving the LBMSW model.…”

Section: Related Workmentioning

confidence: 99%

Real-Time Lattice Boltzmann Shallow Waters Method for Breaking Wave Simulations

Ojeda

Susín

2014

Communications in Computer and Information Science

View full text Add to dashboard Cite

Abstract. We present a new approach for the simulation of surfacebased fluids based in a hybrid formulation of Lattice Boltzmann Method for Shallow Waters and particle systems. The modified LBM can handle arbitrary underlying terrain conditions and arbitrary fluid depth. It also introduces a novel method for tracking dry-wet regions and moving boundaries. Dynamic rigid bodies are also included in our simulations using a two-way coupling. Certain features of the simulation that the LBM can not handle because of its heightfield nature, as breaking waves, are detected and automatically turned into splash particles. Here we use a ballistic particle system, but our hybrid method can handle more complex systems as SPH. Both the LBM and particle systems are implemented in CUDA, although dynamic rigid bodies are simulated in CPU. We show the effectiveness of our method with various examples which achieve real-time on consumer-level hardware.

show abstract

“…Geist et al developed a real-time approach for wave surface generation and attained 25 Hz for grid of size 1024 2 [19]. Geveler et al developed LBM based approach for simulating laminar flow with free fluid surface on multi-core CPU and many-core GPU processors and reported a factor of 8 speed-up on GPU code with respect to multi-threaded CPU code [20]. Gladkov et al reported direct simulation Monte Carlo (DSMC) for solving Boltzmann equations on per particle basis (to solve rarefied fluid flow problems) by using GPU to obtain speed-up by a factor of 65 over serial implementation.…”

Section: Fluid-rigid Body Interaction Simulationmentioning

confidence: 99%

GPU based generation of state transition models using simulations for unmanned surface vehicle trajectory planning

Thakur

Švec

Gupta

2012

Robotics and Autonomous Systems

View full text Add to dashboard Cite

This paper describes GPU based algorithms to compute state transition model for unmanned surface vehicles (USVs) using 6 degree of freedom (DOF) dynamics simulations of vehicle-wave interaction. State transition model is a key component of Markov Decision Process (MDP), which is a natural framework to formulate the problem of trajectory planning under motion uncertainty. USV trajectory planning problem is characterized by the presence of large and somewhat stochastic forces due to ocean waves, which can cause significant deviations in their motion. Feedback controllers are often employed to reject disturbances and get back on the desired trajectory. However, the motion uncertainty can be significant and must be considered in the trajectory planning to avoid collisions with the surrounding obstacles. In case of USV missions, state transition probabilities need to be generated on-board, to compute trajectory plans that can handle dynamically changing USV parameters and environment (e.g., changing boat inertia tensor due to fuel consumption, variations in damping due to changes in water density, variations in sea-state, etc.). The 6 DOF dynamics simulations reported in this paper are based on potential flow theory. We also present a model simplification algorithm based on temporal coherence and its GPU implementation to accelerate simulation computation performance. Using the techniques discussed in this paper we were able to compute state transition probabilities in less than 10 minutes. Computed transition probabilities are subsequently used in a stochastic dynamic programming based approach to solve the MDP to obtain trajectory plan. Using this approach, we are able to generate dynamically feasible trajectories for USVs that exhibit safe behaviors in high sea-states in the vicinity of static obstacles.

show abstract

Lattice-Boltzmann Simulation of the Shallow-Water Equations with Fluid-Structure Interaction on Multi- and Manycore Processors

Cited by 20 publications

References 9 publications

A simulation suite for Lattice-Boltzmann based real-time CFD applications exploiting multi-level parallelism on modern multi- and many-core architectures

A simulation suite for Lattice-Boltzmann based real-time CFD applications exploiting multi-level parallelism on modern multi- and many-core architectures

Real-Time Lattice Boltzmann Shallow Waters Method for Breaking Wave Simulations

GPU based generation of state transition models using simulations for unmanned surface vehicle trajectory planning

Contact Info

Product

Resources

About