Open AccessCPU utilization prediction method based on composite model
Author(s) -
Benran Hu,
Wuzhuo Peng,
Yanjun Li,
Jiyuan Ren,
Yiqun Li
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1693/1/012112
Subject(s) - computer science , central processing unit , multi core processor , decomposition , artificial neural network , artificial intelligence , data mining , parallel computing , operating system , ecology , biology
The business system server under the big data environment is always faced with a large number of data services, which makes the performance of its core hardware CPU face a serious test. Therefore, the prediction of server CPU utilization is of great significance in server resource allocation. In response to this situation, a method of server CPU utilization prediction based on ARMA-BiLSTM composite model is proposed. In this method, the original CPU utilization data is decomposed by wavelet decomposition, and the trend item and the detail item of CPU utilization data are obtained. Based on the Bi-LSTM neural network model, the trend item is modeled and predicted, and the detail item is modeled and predicted using the ARMA model. The prediction results of the two are added together to obtain the final prediction result. Through experimental verification, the proposed prediction model has higher accuracy than traditional models.