面向云计算的主成分分析多变量局域预测模型  

Multivariate-local forecasting model of principal component analysis for cloud computing

作　　者：李洪安[1] 康宝生[1] 张婧[1] 佟建锋[1] 

机构地区：[1]西北大学信息科学与技术学院,西安710127

出　　处：《计算机应用研究》2012年第11期4170-4175,共6页Application Research of Computers

基　　金：西北大学研究生创新教育项目(10YSY02;08YKC17)

摘　　要：针对云平台无法从单变量负荷序列中获取完整预测信息的问题,提出了一种基于主成分分析的多变量局域预测模型并应用于云计算底层资源的预测中。利用主成分分析法综合考虑多种底层资源间的影响关系,确定多变量相空间的嵌入维数,并与局域预测法相结合,由此建立多变量局域预测模型。仿真实验表明,基于主成分分析的多变量局域预测模型的预测精度高于单变量局域预测模型,是面向云计算底层资源预测的一种有效方法。To solve the problems that cloud computing system can not get enough forecasting information from the univariate load sequence,this paper proposed the multivariate-local forecasting model based on principal component analysis,and applied it in the forecasting of the underlying resource for cloud computing.Using of principal component analysis method to consider the relationship between the underlying resources,it determined the embedding dimension of multivariate phase space and combine with the local forecasting method.The simulation results show that the multivariate-local forecasting model based on principal component analysis can provide more precise of forecasting than the univariate-local forecasting model.Thus,the multivariate-local forecasting model is demonstrated to be efficient for the forecasting of the underlying resource for cloud computing.

关 键 词：云计算 多变量相空间 主成分分析 局域预测模型 

分 类 号：TP393.06[自动化与计算机技术—计算机应用技术]