Optimizing Thermal Performance of Data Centers with Novel Local Partition Configurations

2018 2nd International Conference on Power and Energy Engineering

Lu, Hongjie^1 ; Zhang, Zhongbin^1

School of Energy and Mechanical Engineering, Nanjing Normal University, Nanjing, China^1

关键词: Air distribution;    Economic problems;    Heat density;    Indoor environment;    Local overheating;    Partition model;    Reliability problems;    Thermal Performance;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/192/1/012068/pdf
DOI  :  10.1088/1755-1315/192/1/012068

来源: IOP

【 摘 要 】

The indoor environment of data center today has differed from the legacy data centers from the heat density to the air distribution. The energy consumption of cooling system has been taking up a large percent of the total energy of data centers. The reliability problem caused by local overheating and economic problem by undercooling attract the attention. novel local partition configurations are proposed in this paper to optimizing the air distribution and thermal performance of data centers. A basic model and 6 partition models have been simulated to analyze the effectiveness of the novel configurations. The result proves that the configurations with local enclosed partition at tile No.2 show great performance in eliminating the hot spots and undercooling problems.