With the rapid development and application of artificial intelligence (AI) technology, data centers, as important infrastructure in the digital era, have attracted increasing attention to their energy consumption and operating costs. IDC's latest report reveals the role of AI in driving the growth of data center energy consumption and predicts that data center facility spending will increase against the backdrop of rising power costs.
The growth of AI workloads has led to a surge in demand for computing power, which is directly reflected in the growth of data center energy consumption. IDC predicts that AI data center energy consumption will reach 146.2 billion kilowatt hours (TWh) by 2027, showing a significant increase in the proportion of AI workloads in total data center power consumption. At the same time, global data center power consumption is expected to more than double between 2023 and 2028, growing at a compound annual growth rate of 19.5% to reach 857 billion kilowatt hours (TWh) by 2028.
The rising cost of electricity poses a significant challenge for data center operators. Electricity is the largest portion of data center operating costs, accounting for 46% of total enterprise data center spend and 60% of service provider data centers. With the rise in electricity prices, the operating costs of data centers are expected to increase significantly. To meet this challenge, data center operators are looking for ways to improve energy efficiency and adopt renewable energy.
Figure: Electricity-eating Mania AI is driving the move towards green energy
Tencent's distributed new energy microgrid project at the Tianjin High-tech Cloud Data Center is a good example. The project uses solar power to generate electricity with a total installed capacity of 10.54 megawatts, and is expected to save 12 million kilowatt-hours of electricity per year, equivalent to the annual electricity consumption of 6,000 households. In addition, Tencent has introduced an AI energy optimization system that uses machine learning to predict power demand and power generation to improve energy management.
The green transformation of data centers not only helps reduce energy consumption and operating costs, but also has great significance for achieving global carbon neutrality goals. Chinese companies' overseas investment in renewable energy has also shown positive attitudes and actions. For example, Tencent will install more than 32 MW of spontaneously generated new energy in 2023, and in 2024, it will contract more than 1.3 billion kWh of green power purchases, and the annual electricity consumption of green power in its own data centers has reached 54%.
In addition, the Chinese government has become aware of the problem of data center energy consumption and has introduced a series of policies to promote the development of green data centers. The "Special Action Plan for Green and Low-Carbon Development of Data Centers" proposes that by the end of 2025, the layout of data centers across the country will be more reasonable, the overall shelf rate will not be less than 60%, the average power utilization efficiency will be reduced to less than 1.5, and the average annual growth rate of renewable energy utilization will be 10%.
In summary, the development of AI technology is driving the growth of data center energy consumption, while the rising cost of electricity is driving data center operators to seek more efficient energy management and the use of renewable energy. Through technological innovation and the adoption of green energy, the data center industry is moving in a more sustainable direction.
