Multi-GPU performance of incompressible flow computation by lattice Boltzmann method on GPU cluster

Xian, Wang; Aoki, Takayuki

doi:10.1016/j.parco.2011.02.007

Cited by 61 publications

(48 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A popular architecture is the GPU cluster using a mixed MPI and GPU threaded model, for which a number of implementations of LB hydrodynamics have been described in the literature [4,24,26,34,37]. Recently, we [15] described an implementation for colloidal suspensions which employed the GPU for lattice-based operations, but retained a small number of colloid-based operations on the CPU.…”

Section: Parallel Implementationmentioning

confidence: 99%

Large Colloids in Cholesteric Liquid Crystals

2015

View full text Add to dashboard Cite

We describe a coarse-grained Landau-de Gennes model of liquid crystals (LCs) including hydrodynamics based on the Beris-Edwards equations. The model is employed to study the impact of large colloids on the long range LC defect structure in the cholesteric LC blue phases. 'Large' here means that the particle size is comparable to the cholesteric pitch, the length scale on which the LC order undergoes a helical twist. We investigate the case of a single particle, with either normal or degenerate planar anchoring, placed initially in an equilibrium blue phase LC. It is found that in some cases, well defined steady disclination structure emerges at the particle surface, while in other cases no clear steady state is reached in the simulations, and disclination reorganisation appears to proliferate through the bulk LC. These systems are of potential interest in the context of using LCs to template self-assembly of colloid structure, e.g., for opto-electronic devices. Computationally, we demonstrate a parallel approach using mixed message-passing and threaded model on graphical processing units allows effective and efficient progress for this problem.

show abstract

Section: Parallel Implementationmentioning

confidence: 99%

Large Colloids in Cholesteric Liquid Crystals

2015

View full text Add to dashboard Cite

show abstract

“…The researchers started to develop GPUs-based MD simulation programs because of GPUs' significantly computational efficiency, relatively low prices and energy consumption, and all the performance of their implementations increased by tens of times [20][21][22]. Later, several implementations of MD [23] and lattice Boltzmann method (LBM) [24,25] have supported running simulations on multiple GPU devices. MD packages, such as NAMD, LAMMPS and Gromacs, have supported GPU acceleration.…”

Section: Introductionmentioning

confidence: 99%

The aggregation and diffusion of asphaltenes studied by GPU-accelerated dissipative particle dynamics

Wang

Wen

2014

Computer Physics Communications

View full text Add to dashboard Cite

“…With adequate computational resources, distributed computing can be highly efficient to handle many large scale scientific problems [27][28][29][30][31][32] . For instance, Wijerathne et al [33] used a cluster of workstations to simulate the seismic damage of buildings in Tokyo.…”

Section: Introductionmentioning

confidence: 99%

A computational framework for regional seismic simulation of buildings with multiple fidelity models

Law

2016

Advances in Engineering Software

View full text Add to dashboard Cite

Multi-GPU performance of incompressible flow computation by lattice Boltzmann method on GPU cluster

Cited by 61 publications

References 20 publications

Large Colloids in Cholesteric Liquid Crystals

Large Colloids in Cholesteric Liquid Crystals

The aggregation and diffusion of asphaltenes studied by GPU-accelerated dissipative particle dynamics

A computational framework for regional seismic simulation of buildings with multiple fidelity models

Contact Info

Product

Resources

About