10 Data Center Facility Trends for 2025
From 2010 to 2019, the data center industry experienced a spectacular decade, with data centers evolving from ICT equipment rooms to cloud data centers. The rapid development of new tech like artificial intelligence, cloud computing, big data, and 5G will create a new golden era of data-driven productivity. And market demand for data centers will increase dramatically.
At the same time, the data center world isn’t without challenges: it’s difficult to get construction resources, they take a long time to build, and they use a ton of energy. On top of that, architecture flexibility and O&M are key issues that need heavy consideration.
In light of the last decade’s development trajectory, here are 10 trends that we predict will define data center facilities for 2025.
Trend 1: High Density
CPU and server capacity keep increasing alongside gains in IT computing capacity. As demand for AI applications increases, the importance of AI computing power also increases. To balance efficiency and costs, data centers will become high density. Currently, the average power capacity in a data center is 6 to 8 kW/rack. It’s anticipated that a power density of 15 to 20 kW/rack will be dominate data centers by 2025.
Trend 2: Scalable Architecture
Generally, the lifecycle of IT devices is 3 to 5 years, and the power density doubles every 5 years. However, the lifecycle of data center infrastructure is 10 to 15 years. The data center facility will support IT device evolution for 2 to 3 generations. It demands scalable expansion and phased investment for optimal CAPEX in the data center lifecycle. In addition, data centers must support the hybrid deployment of IT devices with different power densities because of the diverse range of IT services running there.
Trend 3: Green
Currently, the power consumption of data centers accounts for 3% of the world’s total power consumption. It’s estimated that the total power consumption of data center will reach more than 1,000 TW by 2025 – the equivalent of 1 trillion tons of coal. Saving energy, cutting emissions, and lowering OPEX are big challenges that involves lowering the figure for power usage effectiveness (PUE) and building green data centers. We can expect to see clean energy, waste heat recovery, and reused resources such as energy, land, water, and materials. It is estimated that the average PUE of new data center in China will drop to 1.1 in the next five years, down from 1.67 in 2019.
Trend 4: Quick Deployment
Data centers must be rolled out quickly, as Internet services usually create increases in data and traffic, which ramps up requirements on the service side sharply. Moreover, data centers have changed from a support system to a production system. Faster rollout means faster benefits. The typical TTM of a data center is 9 to 12 months, which is expected to drop to less than 6 months by 2025.
Trend 5: Full Digitalization & AI-enablement
Data center facilities will become digital and intelligent. With continual improvements to IoT and AI technologies, data centers will gradually evolve from single-domain digitalization in terms of O&M, energy saving, and operations to full-lifecycle digitalization and autonomous in terms of planning, construction, O&M, and optimization. AI will be widely applied.
Trend 6: Full Modularization
More data centers will be constructed in full modular mode to mitigate slow construction and high initial investment costs. Modular design will evolve from component modularization to architecture modularization and equipment room modularization to finally achieve full modularization of data center. The full modular design will enable fast deployment, flexible capacity expansion, simple O&M, and high energy efficiency.
Trend 7：Simplified power supply architecture, with lithium batteries replacing lead-acid batteries
The power supply and distribution system of a traditional data center is complex and fragmented, generates a large footprint, and it’s difficult to locate faults. A simplified power supply architecture will reduce power conversion times, shorten the power supply distance and footprint, and improve the space utilization rate and system energy efficiency. Compared with lead-acid batteries, lithium batteries have advantages in terms of footprint and service life. As the cost of lithium batteries decreases, lithium batteries will be widely used in data centers in the future.
Trend 8: Convergence of liquid cooling and air cooling, more indirect evaporative Cooling & and less chilled water cooling
The complex architecture of the chilled water cooling system slows deployment and complicates O&M – both of which problems can be solved by an indirect evaporative cooling system with modular architecture. Using natural cooling resources will greatly reduce the power used by the cooling system. In areas with suitable climates, the chilled water system will gradually be replaced by indirect evaporative cooling system.
GPU and NPU applications generate more high-density scenarios, and liquid cooling systems will be increasingly adopted. However, some storage and computing services are still in low-density scenarios. To quickly adapt to uncertain IT service requirements in the future, the cooling solution must be compatible with the air cooling system and liquid cooling system.
Trend 9: Dynamic Link between Bits and Watts
Reducing PUE doesn’t mean that the overall energy consumption of the data center is optimal. Instead of focusing on the data center energy facilities, the energy consumption of the data center needs to be evaluated and optimized as a whole. Innovations in facilities, IT, chipsets, data, and cloud will enable bits and watts will work in tandem to achieve dynamic energy saving and optimal energy efficiency system-wide.
Trend 10: Trustworthiness
As data center facilities become smarter, network security threats will multiply. The data center must have six features: resilience, security, privacy, safety, reliability, and the availability to prevent attacks and threats from environments and malicious personnel, including network intrusion threats.