Study on the Prediction Models of Temperature and Energy by using DCIM and Machine Learning to Support Optimal Management of Data Center

摘要

Data centers (DCs) are becoming increasingly important. Accordingly, highly efficient and reliable operation and management of DCs have been required. Conventional DCs operate information and communication technology (ICT) and facility management (FM) systems separately, which could lead to inefficient management. Nowadays, DCs have been focusing on the data center infrastructure management (DCIM) system, in which ICT equipment, power equipment, and other heating, ventilation, and air conditioning (HVAC) devices can be managed in an integrated manner. In this paper, we propose a method of designing rack and ICT placement as an example of initiatives that realize proper temperature management and energy saving effects using DCIM and machine learning (ML). Then models for predicting the temperature energy after enviroment chages by using machine learning are developed, and the results of verification and effectiveness in the verification room are reported.