The 33rd China International Exhibition on Electric Power Equipment and Technology
Shanghai International Energy Storage Technology Application Expo / Hydrogen Energy Expo
A green data centre is a facility for housing computer systems and associated components — servers, storage, networking — that is designed and operated to minimise energy consumption, water usage, and carbon emissions. Key green data centre strategies include: renewable energy procurement (power purchase agreements for wind and solar, or on-site generation); high-efficiency cooling systems (free cooling using outside air or water, liquid cooling for high-density servers, AI-optimised cooling control); high power usage effectiveness (PUE) targets (best-in-class data centres achieve PUE below 1.2, compared to the industry average of 1.5–1.6); waste heat recovery for district heating or industrial processes; and battery energy storage for demand management and backup power. China's data centre industry is under increasing pressure to reduce energy consumption and carbon emissions, driven by government policy (the 'East Data West Computing' strategy) and corporate sustainability commitments from major cloud providers.
5 Key Questions About Green Data Centre
Power Usage Effectiveness (PUE) is the ratio of total data centre energy consumption to IT equipment energy consumption. A PUE of 1.0 would mean all energy goes to IT equipment with zero overhead; a PUE of 2.0 means equal energy is used for cooling, power distribution, and other overhead as for IT equipment. Best-in-class hyperscale data centres (Google, Microsoft, Meta) achieve PUE below 1.1 using advanced cooling and power distribution designs. China's national standard GB/T 32910 sets PUE targets for new data centres: below 1.4 for general data centres and below 1.25 for large data centres in northern China where free cooling is available.
Green data centres use a range of advanced cooling technologies: air-side economisers using outside air for free cooling when temperatures are suitable; water-side economisers using cooling towers or dry coolers; direct liquid cooling (DLC) circulating liquid directly to server components for high-density racks; immersion cooling submerging servers in dielectric fluid for maximum efficiency; and AI-powered cooling control systems that continuously optimise cooling equipment operation based on IT load, weather, and energy prices. Liquid cooling is becoming increasingly important as AI server power densities exceed 50–100 kW per rack, far beyond the capability of traditional air cooling.
Data centres integrate with renewable energy through: power purchase agreements (PPAs) for wind and solar energy; on-site solar generation on rooftops and parking structures; co-location with renewable energy projects (particularly in China's 'East Data West Computing' zones in Guizhou, Inner Mongolia, and Gansu where renewable energy is abundant); battery energy storage for demand management and renewable energy time-shifting; and participation in demand response programmes that shift flexible computing workloads to periods of high renewable generation. Major Chinese cloud providers including Alibaba Cloud, Tencent Cloud, and ByteDance have committed to 100% renewable energy targets.
China's 'East Data West Computing' (东数西算) strategy, launched in 2022, aims to shift data centre construction from energy-constrained eastern cities to western regions with abundant renewable energy and land. Eight national computing hubs have been designated in Guizhou, Inner Mongolia, Gansu, Ningxia, Xinjiang, Beijing-Tianjin-Hebei, Yangtze River Delta, and Guangdong-Hong Kong-Macao Greater Bay Area. Western hubs benefit from lower land costs, cooler climates (enabling free cooling), and abundant wind and solar energy. The strategy aims to improve national data centre energy efficiency while reducing carbon emissions from the growing digital economy.
AI and machine learning are transforming data centre energy management: AI-powered cooling control systems (pioneered by Google DeepMind) continuously optimise cooling equipment settings based on IT load, weather, and energy prices, achieving 30–40% cooling energy reductions. Predictive maintenance algorithms analyse sensor data to identify equipment approaching failure before outages occur. Workload scheduling systems shift flexible computing tasks to periods of low energy cost or high renewable generation. AI-powered power distribution management optimises UPS and power conversion efficiency. These AI applications are moving data centres from static, over-provisioned designs to dynamic, demand-responsive facilities.
Key Takeaways
Green data centres are a critical intersection of the digital economy and the energy transition, as the explosive growth of AI and cloud computing creates massive new electricity demand that must be met with clean energy. China's data centre industry is under strong policy pressure to improve energy efficiency and adopt renewable energy. EP Shanghai showcases the power equipment, cooling systems, energy storage, and management platforms that enable green data centre development.
Firefox
Chrome
Safari
QQ
Edge