摘要
近年来,统一计算设备架构(CUDA)的提出和图形处理器(GPU)快速提升的并行处理能力和数据传输能力,使得基于CUDA的GPU通用计算迅速成为一个研究热点。针对含有大规模分子动力学模拟的热力学量提取效率低下的问题,提出了分子动力学模拟的热力学量提取的新方法,利用CUDA设计了并行算法,实现了利用GPU加速分子动力学模拟的热力学量提取。实验结果表明,与基于CPU的算法相比,GPU可以提高速度500倍左右。
Recently, the compute unified device architecture (CUDA) and the rapidly promoting ability of data-parallel processing and data transferring of graphic processing units (GPU) make an attractive area for general purpose computation based on CUDA. As the extraction of thermodynamic quantities in large scale molecular dynamics simulation was relative low, this paper proposed a new approach to the extraction of thermodynamics quantities of molecular dynamics simulations, and used the CUDA to design and implement parallel algorithm. The results indicate that GPU achieves very high speedup of up to near 500 folds compared with CPU-based algorithm.
出处
《计算机应用研究》
CSCD
北大核心
2010年第5期1820-1822,共3页
Application Research of Computers
基金
国家教育部高校行动计划智能科学与技术项目(2004XD-03)