Load balancing n-body simulations with highly non-uniform density
Load balancing n-body simulations with highly non-uniform density
复制标题
具有高度不均匀密度的负载平衡 n 体模拟
DOI：
--
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
N. Amato
中科院分区：
文献类型：
--
作者：
Olga Pearce;T. Gamblin;B. Supinski;T. Arsenlis;N. Amato
N体方法模拟粒子（或物体）系统的演化。它们对于分子动力学、天体物理学和材料科学等不同领域的科学研究至关重要。N体方法的大多数负载平衡技术使用粒子计数来近似计算工作。这种近似是不准确的，特别是对于具有高密度变化的系统，因为N体模拟中的功与粒子密度成比例，而不是粒子计数。在本文中，我们证明了现有的技术在颗粒密度高度不均匀时，在规模上表现不佳，我们提出了一种负载平衡技术，有效地分配负载的相互作用，而不是颗粒。我们使用自适应采样来创建一个更适合分区的均匀工作分布，并减少分区开销。我们实施和评估我们的方法上的Barnes-Hut算法和大规模的位错动力学应用程序，帕拉迪斯。我们的方法实现了高达26%的Barnes-Hut和18%的帕拉迪斯的整体性能的改善。
N-body methods simulate the evolution of systems of particles (or bodies). They are critical for scientific research in fields as diverse as molecular dynamics, astrophysics, and material science. Most load balancing techniques for N-body methods use particle count to approximate computational work. This approximation is inaccurate, especially for systems with high density variation, because work in an N-body simulation is proportional to the particle density, not the particle count. In this paper, we demonstrate that existing techniques do not perform well at scale when particle density is highly non-uniform, and we propose a load balance technique that efficiently assigns load in terms of interactions instead of particles. We use adaptive sampling to create an even work distribution more amenable to partitioning, and to reduce partitioning overhead. We implement and evaluate our approach on a Barnes-Hut algorithm and a large-scale dislocation dynamics application, ParaDiS. Our method achieves up to 26% improvement in overall performance of Barnes-Hut and 18% in ParaDiS.