计算机应用与软件2023，Vol.40Issue(12)：48-55,100,9.DOI:10.3969/j.issn.1000-386x.2023.12.008
基于改进的Stacking集成模型的容器云负载预测研究
CONTAINER CLOUD LOAD FORECASTING BASED ON IMPROVED STACKING ENSEMBLE MODEL
摘要
Abstract
With the rapid development of container cloud,the volume of business increases rapidly,so it is a reasonable way to forecast the future trend of container resource utilization and allocate resources in advance to improve the utilization rate of resources and reduce the waste of resources.In order to realize reasonable prediction of container cloud resources,a cloud resource load prediction model based on the improved Stacking ensemble method is proposed.The first stage of the model was to set up a base learner to select the features of cloud resource load data and reduce the complexity of the features of the data set.The second stage was to use the DBN model improved by GA-BP neural network,which named DBN-GA-BP,to carry out integrated prediction of the feature selection data in the first stage.The experimental results show that the model has higher prediction accuracy compared with the single model and the unimproved Stacking model.关键词
Stacking集成模型/遗传算法/深度信念网络/云资源/资源预测Key words
Stacking ensemble model/Genetic algorithm/Deep belief nets/Cloud resource/Resource prediction分类
信息技术与安全科学引用本文复制引用
梁荣华,谢晓兰,翟青海,张启明..基于改进的Stacking集成模型的容器云负载预测研究[J].计算机应用与软件,2023,40(12):48-55,100,9.基金项目
国家自然科学基金项目(61762031) （61762031）
广西科技重大专项(桂科AA19046004) （桂科AA19046004）
广西重点研发项目(桂科AB18126006). （桂科AB18126006）
