New multi-GPU implementation for smoothed particle hydrodynamics on heterogeneous clusters
New multi-GPU implementation for smoothed particle hydrodynamics on heterogeneous clusters
复制标题
DOI：
10.1016/j.cpc.2013.03.008
复制
发表时间：
2013-08-01
影响因子：
6.3
通讯作者：
Gomez-Gesteira, M.
中科院分区：
文献类型：
--
作者：
Dominguez, J. M.;Crespo, A. J. C.;Gomez-Gesteira, M.
提出了一种利用中央处理器（cpu）和图形处理器（gpu）异构集群的大规模并行SPH方案。新的实现源自单gpu的dualspphysics代码，以前被证明是强大的，稳定的和准确的。不同并行编程语言的组合不仅可以利用一个设备（CPU或GPU），还可以利用不同机器的组合。设备之间的通信使用改进的消息传递接口（MPI）实现，该实现解决了MPI的一些众所周知的缺点，例如包括动态负载平衡和重叠的数据通信和计算任务。在不同粒子数量和不同gpu数量下，分析了新dualspphysics代码的效率和可扩展性（强缩放和弱缩放）。最后，提出了一个超过10(9)个粒子的应用程序，以显示代码处理模拟的能力，否则需要大型CPU集群或超级计算机。(C) 2013 Elsevier B.V.版权所有
A massively parallel SPH scheme using heterogeneous clusters of Central Processing Units (CPUs) and Graphics Processing Units (GPUs) has been developed. The new implementation originates from the single-GPU DualSPHysics code previously demonstrated to be powerful, stable and accurate. A combination of different parallel programming languages is combined to exploit not only one device (CPU or GPU) but also the combination of different machines. Communication among devices uses an improved Message Passing Interface (MPI) implementation which addresses some of the well-known drawbacks of MPI such as including a dynamic load balancing and overlapping data communications and computation tasks. The efficiency and scalability (strong and weak scaling) obtained with the new DualSPHysics code are analysed for different numbers of particles and different number of GPUs. Last, an application with more than 10(9) particles is presented to show the capability of the code to handle simulations that otherwise require large CPU clusters or supercomputers. (C) 2013 Elsevier B.V. All rights reserved.