Accurate server energy consumption model can predict data center energy consumption and provide important basis for resource scheduling method.Considering the problems of high dimension of server energy consumption characteristics and difficulty in judging redundant features,this paper analyzes the relationship between features and features and between features and target values of Pearson correlation coefficient,and gives the judgment criteria for server energy consumption redundant features.On this basis,a server energy consumption feature selection algorithm based on redundancy analysis is proposed Ex-perimental results show the effectiveness of the proposed energy consumption feature selection algorithm in server energy con-sumption model construction.
关键词
云数据中心/服务器/冗余分析/特征选择/能耗预测
Key words
cloud data center/server/redundancy analysis/feature selection/energy consumption prediction
维普
万方数据