English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Open

Presents the pdf analysis as premium feature

PDF Analysis

Insights

Presents the summarisation as premium feature

Summarisation

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

The Lattice Boltzmann method (LBM) for solving fluid flow is naturally wellsuited to an efficient implementation for massively parallel computing, due tothe prevalence of local operations in the algorithm. This paper presents andanalyses the performance of a 3D lattice Boltzmann solver, optimized for thirdgeneration nVidia GPU hardware, also known as `Kepler'. We provide a review ofprevious optimisation strategies and analyse data read/write times fordifferent memory types. In LBM, the time propagation step (known as streaming),involves shifting data to adjacent locations and is central to parallelperformance; here we examine three approaches which make use of differenthardware options. Two of which make use of `performance enhancing' features ofthe GPU; shared memory and the new shuffle instruction found in Kepler basedGPUs. These are compared to a standard transfer of data which relies instead onoptimised storage to increase coalesced access. It is shown that the moresimple approach is most efficient; since the need for large numbers ofregisters per thread in LBM limits the block size and thus the efficiency ofthese special features is reduced. Detailed results are obtained for a D3Q19LBM solver, which is benchmarked on nVidia K5000M and K20C GPUs. In the lattercase the use of a read-only data cache is explored, and peak performance ofover 1036 Million Lattice Updates Per Second (MLUPS) is achieved. Theappearance of a periodic bottleneck in the solver performance is also reported,believed to be hardware related; spikes in iteration-time occur with afrequency of around 11Hz for both GPUs, independent of the size of the problem.

Memory transfer optimization for a lattice Boltzmann solver on Kepler architecture nVidia GPUs