Long-term Energy Consumption Forecasting for Data Center Industry in China

The number of internet data centers increases rapidly over the past years, which results in a significant incensement on energy demand. In order to achieve a precise forecast of the data center electricity consumption in the next few years, a scale prediction model for data centers based on Bass model is proposed in this paper. We provide a forecast for the long-term energy consumption in the data centers based on the proposed model while accounting the difference between the existing and planning data centers. The simulation results show that data centers in China will consume more than 400 billion kWh of electricity in 2030, accounting for 4% of the national electricity consumption, and the growth rate will slow down at the same time. We further investigate the impact of implementing advanced technologies on cooling systems and servers. The results show that the decrease of PUE values will effectively reduce electricity consumption, and the increase of server utilization ratios will improve energy efficiency for data centers.