A parallel fast neighbor searching algorithm for particle-based methods on CPU and GPU architectures in multi-scale simulation
Particle-based methods are widely applied in the resolving of complex multi-scale physical phenomena in various science and engineering areas.In order to handle the challenge of increasing com-putational complexity and declining concurrency for the pair-wised particle searching procedure in mas-sive multi-scale particle-based simulations,a new parallel fast neighbor searching algorithm,which fea-tures high-concurrency and low memory footprint,is developed and demonstrated on both many-core CPU and GPU architectures.An inter-level interaction strategy based on the concept of hierarchical nes-ted data structure is proposed to resolve the issue of racing condition in cross-level particle search.An asymmetric mapping method is developed to eliminate the full mapping of particles on each level,which reduces the memory consumption.A set of numerical experiments show that,the proposed algorithm can handle multi-scale problems with particle volume ratio up to 10s.Compared with traditional algo-rithm,the proposed algorithm can achieve 2x～8x speedups and lower memory consumption.The GPU-based implementation of the algorithm achieves state-of-the-art computational efficiency.
国家科技期刊平台
NSTL
万方数据
