Task-Based Parallelism with OpenMP: a case study with DL_POLY_4

Task-Based Parallelism with OpenMP: a case study with DL_POLY_4
复制标题

OpenMP 基于任务的并行性:DL_POLY_4 的案例研究

DOI:
10.1080/08927022.2019.1606424
复制
发表时间:
2019
影响因子:
2.1
通讯作者:
Chalk A
Chalk A
中科院分区:
化学4区
文献类型:
--
作者:
Chalk A

文献摘要

参考文献

相似文献

当执行负载平衡复杂的计算时,动态负载平衡变得越来越必要。在本文中,我们研究了其中一种方法,即基于任务的并行性。许多库都实现了基于任务的并行性,但是,在本文中,我们研究了 OpenMP 标准和实现,并将其应用于经典分子动力学代码 DL_POLY_4,重点关注在许多模拟运行中占计算大部分的两个体积力计算。我们的结果显示使用 OpenMP 任务具有合理的性能,但是,其他库(例如 OmpSs 或 StarPU)中可用的一些扩展可能有助于解决类似于分子动力学的问题,在分子动力学中,避免任务之间的竞争条件可能会产生大量的调度开销。
When performing computations where load balancing is complex, dynamic load balancing is becoming increasingly necessary. In this paper, we examine one of these methods, Task-Based Parallelism. Many libraries implement Task-Based Parallelism, however, in this paper we examine the OpenMP standard and implementations, and apply it to the Classical Molecular Dynamics code, DL_POLY_4, focusing on the two body force calculations that make up a large percentage of the compute in many simulation runs. Our results show reasonable performance using OpenMP tasks, however, some of the extensions available in other libraries such as OmpSs or StarPU may help with performance for problems similar to Molecular Dynamics, where avoiding race conditions between tasks can have a substantial scheduling overhead.
成对 Verlet 列表:组合单元列表和 Verlet 列表以提高内存局部性和并行性
DOI: 10.1002/jcc.21945
发表时间: 2012
影响因子: 3
作者:
P. Gonnet
通讯作者: P. Gonnet
DOI: 10.1002/cpe.3132
发表时间: 2012-03
期刊: Concurrency and Computation: Practice and Experience
影响因子: --
作者:
H. Ltaief;Rio Yokota
通讯作者: H. Ltaief;Rio Yokota
Pseudo-Verlet 列表:一种新的、紧凑的邻居列表表示
DOI: 10.1080/08927022.2012.762097
发表时间: 2013
影响因子: 2.1
作者:
P. Gonnet
通讯作者: P. Gonnet
DOI: --
发表时间: 2014
期刊:
影响因子: --
作者:
R. Meyer
通讯作者: R. Meyer