GPU implementations of the bond fluctuation model
GPU implementations of the bond fluctuation model
复制标题
DOI：
10.1016/j.jcp.2011.12.021
复制
发表时间：
2012-04
期刊：
影响因子：
--
通讯作者：
S. Nedelcu;M. Werner;M. Lang;J. Sommer
中科院分区：
文献类型：
--
作者：
S. Nedelcu;M. Werner;M. Lang;J. Sommer
我们提出了两个并行实现的债券波动模型的图形处理器上的性能优于高达50倍的一个因素，相当于单CPU处理器上的实现。第一个算法是在[S. Nedelcu，J.联合Sommer，聚合物网络中的单链动力学：Monte Carlo研究，J.Chem.Phys.130（2009）204902]。在这第一个算法中，我们使用并行区域分解技术，以避免单体碰撞。相比之下，在第二种算法中，我们将每个单体与一个并行过程相关联，其中系统中的所有单体都试图同时移动。在这两种情况下，只有单体移动，导致允许的债券和保持晶格占有被接受。为了验证GPU算法的正确性，我们模拟了单分散聚合物熔体在单体数密度0.5，并比较静态和动态性能与标准的CPU实现。我们发现CPU和GPU的结果之间有很好的一致性，这证明了串行和并行实现的等效性。讨论了较高单体数密度的影响。
We present two parallel implementations of the bond fluctuation model on graphics processors that outperform by a factor of up to 50 times an equivalent implementation on single CPU processor. The first algorithm is a parallelized version of an accelerated MC method published earlier in [S. Nedelcu, J.-U. Sommer, Single chain dynamics in polymer networks: a Monte Carlo study, J. Chem. Phys. 130 (2009) 204902]. In this first algorithm we use the parallel domain decomposition technique to avoid monomer collisions. In contrast, in the second algorithm we associate each monomer with a parallel process, where all monomers in the system are attempted to move simultaneously. In both cases, only monomer moves that result in allowed bonds and preserve lattice occupancy are accepted. To validate the correctness of the GPU algorithms we simulated monodisperse polymer melts at monomer number density 0.5 and compared static and dynamical properties with standard CPU implementations. We found good agreement between the CPU and the GPU results, which demonstrates the equivalence of the serial and parallel implementations. The influence of higher monomer number density is discussed.