非线性相关的失效数据联合聚类分析与预测

卢旭¹, 王慧强¹, 吕晓², 冯光升¹, 林俊宇¹

1.哈尔滨工程大学计算机科学与技术学院;2.海军工程大学电子工程学院



摘要:

为了实现对高性能计算机系统大规模失效数据的自动分析和预测,引入了基于信息论的联合聚类思想提取非线性相关失效数据对象.根据失效特征非线性相关性对失效数据进行归类,并给出用于聚类的失效特征标签的定义,以此为基础提出以互信息熵作为相似性度量的非线性相关失效数据联合聚类算法,并从理论上论述了算法的收敛性和局部最优性.实验结果显示联合聚类分析算法具有良好的运行性能,成功聚类出非线性相关的失效数据对象,验证了联合聚类分析方法用于失效预测的有效性.

关键词:  失效分析  失效预测  非线性相关性  联合聚类

DOI：10.11918/j.issn.0367-6234.2011.03.017

分类号:TP311.13

基金项目:国家高技术研究发展计划资助项目(2007AA01Z401);国家自然科学基金资助项目(90718003,60973027)



Nonlinearly correlated failure data Co-clustering analysis and prediction

LU Xu¹, WANG Hui-qiang¹, Lü Xiao², FENG Guang-sheng¹, LIN Jun-yu¹

1.College of Computer Science and Technology,Harbin Engineer University,150001 Harbin,China;2.College of Electronic Engineering,Naval Engineer University,430033 Wuhan,China)

Abstract:

To realize automatic analysis and prediction large-scale failure data for high-performance computer systems,an information-theoretic Co-clustering method for large scale nonlinearly correlated failure data was proposed.The failure data were sorted according to the nonlinear correlation of failure features and the failure feature signatures were defined.Then a nonlinearly correlated failure data Co-clustering algorithm was proposed and measured using mutual information entropy.Moreover,the convergence and local optimality of Co-clustering algorithm were proved theoretically.Experimental results on labeled failure data showed that the Co-clustering analysis algorithm outperformed other clustering analysis algorithms and the proposed Co-clustering analysis method had the features of rationality and effectiveness for discovering the nonlinearly correlated failure patterns and could help to predict underlying failures.

Key words:  failure analysis  failure prediction  nonlinear correlation  Co-clustering