Study on Data Center Optimal Management by using DCIM Part5 Temperature Prediction Model after stopping CRAC in Server Room
DCIMを用いたデータセンタにおける最適マネジメントに関する研究 (第五報)機械学習を用いた空調機停止後におけるサーバルームの温度変化予測モデル

Kosuke Sasakura
2020 Techinical Papers of Annual Meeting the Society of Heating,Air-conditioning and Sanitary Engineers of Japan  
Data Center (DC) is increasingly important in recent years. Then, DC has been required efficient and reliable operation and management. So DCIM which can manage ICT equipment, power equipment, and HVAC system, etc. in DC is focused. In this study, we proposed the method to predict rack intake temperature of each rack after stopping computer room air conditioner (CRAC) in server room. Then the prediction model based on machine learning using DCIM information was constructed and the effectiveness was confirmed.
doi:10.18948/shasetaikai.2020.3.0_37 fatcat:vi47brnrxzg3zbk3flz2avaiy4