AI places significant demands on data centre power management. Future-proofing prime power, efficient backup load-stepping, and renewable fuels are key components.
“Increased demand for high-power computing creates a significant price disparity between new data centres and legacy facilities. Many existing data centres lack the infrastructure to handle these demanding workloads, further limiting their appeal.” CBRE North America Data Center Trends Report H12024
With no sign of a let-up in the phenomenal pace of AI infrastructure investment, it is vital for data centre operators to stay competitive by implementing intelligent and sustainable upgrades, such as enhancing power density capabilities, adopting renewable energy solutions, and investing in resilient infrastructure for AI and HPC readiness.
According to a recent report from IDC, AI data centre capacity is projected to have a CAGR of 40.5% through 2027. AI data centre energy consumption is forecast to grow at 44.7%, reaching 146.2 Terawatt hours (TWh) by 2027. AI workloads will consume a growing proportion of total data centre electricity use.
Given this growth in AI-driven demand, it’s not surprising that, in a recent report, CBRE observed a growing price disparity between data centres built or upgraded with the high-density power demands of AI in mind and existing facilities that still lack adequate infrastructure for AI.
Densities are rising fast and will continue, reaching as high as 100 kW/rack, possibly higher. As an industry, we need to advise our customers on configuring their power infrastructure to position themselves for AI resilience and sustainability. This is not just a matter of environmental responsibility. It is also a strategic necessity for the long-term viability of the ever-growing AI industry.
Putting Prime Power in Place
AI is putting demands on the grid that it’s struggling to manage, especially in established hubs like Dublin, Frankfurt, Amsterdam, Paris and London. Waiting times for power capacity are sometimes several years long, creating major headaches for operators. This is where smart microgrids step in, bridging the gap between data centre operators and grid requirements.
AVK’s recent collaboration with a major data centre operator in Slough illustrates how an effective microgrid can transform a facility’s operational outlook. Our client’s critical data centre required a robust power solution that guaranteed operational continuity, scalability, and sustainability.
AVK proposed and installed a microgrid consisting of 19 state-of-the-art generators, including two medium-sized support generators for additional resilience. These generators were optimised to run on Hydrotreated Vegetable Oil (HVO), which reduces carbon emissions by 90% compared to conventional diesel.
This allowed the client to meet their immediate energy demands while also aligning with sustainability goals, making the data centre a net contributor to the local community with the potential to support national infrastructure.
Backup: The Load-Stepping Challenge
Both existing and new facilities must also pay close attention to their backup solutions, which may not be able to keep up with even smaller increases in power density. As customers decentralise their UPS to increase power densities up towards the 50/100 kW level, they need smooth load-stepping, which means the ability to gradually increase the electrical load on a backup power system, such as a generator.Smooth load-stepping allows backup power systems to handle growing demand without overloading or causing instability.
Understanding Your Power Needs
Sustainable, high-density infrastructure can be assembled piece by piece as long as organisations apply principles like energy efficiency, scalability, and resilience at every step and build for the future.
Microgrid design, load management, and renewable fuels are critical elements to consider as power levels increase to support AI. The market is also evolving quickly—new solutions are being developed that increasingly rely on AI to optimise their effectiveness, making AI part of the solution and the challenge.
Perhaps the most essential principle for tackling these issues is understanding power requirements intimately—from prime power and load-stepping to backup fuel choice.
The journey towards AI readiness is about more than keeping up with technological advancements—it’s about building solutions that are scalable, sustainable, and capable of supporting the industry in the long term.
