摘要
研究大数据环境下网格动态故障检测的方法。大数据来源范围广博,数据类型极复杂;数据的广泛性,资源的高度异构和不同地理上的分布,使网格故障发生成为影响系统应用的主要问题。目前网格故障检测方式,不能满足网格动态故障检测需要。利用"灰色预测理论"的算法,依据动态心跳的原理,设计动态故障检测架构,给出了预测模型;提出了网格动态故障检测方法。实验结果证实是有效的和准确的,提出的动态故障检测算法优于静态故障检测算法,解决了大数据环境下网格动态故障检测问题。
We study the method of dynamic grid fault detection in big data environment. The source range of big data is extensive, the data types are extremely complex; the extensiveness of data,the highly heterogeneous resources and the different geographical distribution make the grid fault occurrence become the major problem affecting system applications. Current grid fault detection mode can not meet the needs of dynamic grid fault detection. W e use a grey prediction theoryr,algorithm and based on the principle of dynamic heartbeat to have designed the dynamic fault detection architecture,and give the prediction model; we propose the dynamic grid fault detection method. Experimental results prove that it is effective and accurate,the proposed dynamic fault detection algorithm outperforms the static fault detection algorithm,it solves the problem of dynamic grid fault detection in big data environment.
作者
李景林
Li Jinglin(College of Humaniyes and Sciences, Guizhou Minzu University, Guiyang 550025 , Guizhou, China)
出处
《计算机应用与软件》
CSCD
2016年第6期51-53,75,共4页
Computer Applications and Software
基金
贵州省科学技术基金项目(黔科合J字[2010]2106号)
关键词
大数据
网格
故障
检测
研究
Big data
Grid
Fault
Detection
Research