《电子技术应用》
您所在的位置:首页 > 其他 > 设计应用 > 基于矩阵嵌套的CESM负载均衡优化方案检索策略
基于矩阵嵌套的CESM负载均衡优化方案检索策略
2022年电子技术应用第1期
董润婷,吴 利,黄建强,王晓英
青海大学 计算机技术与应用系,青海 西宁810016
摘要: 通用地球系统模式(Community Earth System Model,CESM)是一种定量描述气候系统模式变化的数值模式,庞大的科学计算体量已使其成为高性能计算领域的重要研究对象之一。CESM各个气象子模块及组件之间的负载不均衡问题使其计算性能一直不理想,且由于其可用的进程布局方案本身的多样性会导致的巨大检索量,因此通过人工调参用枚举寻找最优布局本身并不现实。为解决这个问题,提出并实践了一种基于矩阵嵌套思想的负载平衡优化方案检索策略帮助完成进程布局优化过程,并介入基于模式并行要求的筛查保证检索结果具有可行性,最终通过实验证明通过这种检索策略搜索获得的最优布局与默认布局相比平均计算性能提升达到47.3%,并在5个节点上实现了1.419的加速比。
中图分类号: TP311
文献标识码: A
DOI:10.16157/j.issn.0258-7998.212426
中文引用格式: 董润婷,吴利,黄建强,等. 基于矩阵嵌套的CESM负载均衡优化方案检索策略[J].电子技术应用,2022,48(1):24-30.
英文引用格式: Dong Runting,Wu Li,Huang Jianqiang,et al. A retrieval strategy of load balancing optimization scheme for CESM based on matrix nesting[J]. Application of Electronic Technique,2022,48(1):24-30.
A retrieval strategy of load balancing optimization scheme for CESM based on matrix nesting
A retrieval strategy of load balancing optimization scheme for CESM based on matrix nesting
Department of Computer Technology and Application,Qinghai University,Xining 810016,China
Abstract: The Community Earth System Model(CESM) is a numerical model to quantitatively describe the change of climate system model, which is one of the most important research objects in the field of high-performance computing because of its huge volume of scientific computing. The load imbalance between each meteorological sub module and component of CESM makes its computing performance unsatisfactory. It is not realistic to retrieve the optimal layout manually by enumerating parameters because the diversity of available process layout schemes will lead to a huge amount of retrieval. In order to solve this problem, this paper proposed and implemented a retrieval strategy based on the matrix-nesting idea of load balancing optimization scheme to help the process layout and intervenes in the screening work based on the parallel requirements of the original model. Finally, the experiment proved that the optimal layout obtained by this search strategy search had a performance improvement of 47.3% compared with the default layout and achieved an acceleration ratio of 1.419 on 5 nodes.
Key words : high performance computing;community earth system model;matrix-nesting;load balancing

0 引言

    在全球变暖的背景下,极端天气事件发生频率显著增加,严重影响社会经济和人民的生命财产安全,天气预报以及对重大气候异常事件预测的重要性日益凸显。通用地球系统模式(Community Earth System Model,CESM)作为一种定量描述气候系统模式变化的数值模式是气候模拟的重要工具,也是全球领先的地球系统模型之一。由于全球气候变化是一个长期且包含各种物理原理的过程,因此使用地球系统模型进行气候变化模拟通常需要对数十年甚至数百年的模型进行整合推演,并且对计算资源的需求量巨大,地球系统模型也因而成为高性能计算在科学计算领域内极为重要的应用之一,但模型的各个模块在CPU上不均匀的负载分布对整个气象模式的计算速度有非常严重的影响。




本文详细内容请下载:http://www.chinaaet.com/resource/share/2000003904




作者信息:

董润婷,吴  利,黄建强,王晓英

(青海大学 计算机技术与应用系,青海 西宁810016)




wd.jpg

此内容为AET网站原创,未经授权禁止转载。