Keeping systems flowing through better fluid management
8 June 2026 
As data demand accelerates, the pressure on cooling systems is intensifying. Continuous operation and rising heat loads mean fluid condition and maintenance strategies now play a critical role in ensuring reliability and efficiency, as Clive Jones, managing director at Global Heat Transfer, explains
According to the International Energy Agency (IEA), data centres could consume up to three per cent of global electricity demand by 2030, driven largely by AI workloads and high-performance
computing growth. At the same time, while data centres enable digital innovation, their escalating energy demand presents a growing sustainability paradox that should not be ignored.
Traditionally, air cooling has been the default approach for data centres. However, higher rack densities and AIdriven processing loads are pushing air systems to their physical limits. Cooling
already accounts for a substantial share of total power draw, typically representing 30–40% of a data centre’s electricity consumption according to the World Economic Forum (WEF). This is second only to the IT equipment itself.
The resource implications extend beyond electricity. For instance, a Washington Post analysis highlighted that generating a single 100-word email using generative AI can require the equivalent
of a bottle of water to manage server temperatures. This illustrates the hidden water footprint behind digital services.
As demand accelerates, these inefficiencies compound. More energy in means more heat out. Unless cooling approaches evolve, operators risk rising PUE (Power Usage Effectiveness) ratios,
increased downtime exposure, and growing scrutiny from regulators and investors alike.
Liquid cooling moves from niche to necessity
Against this backdrop, liquid cooling, including direct-to-chip systems that cool components via cold plates and immersion methods that submerge servers in coolant, are gaining momentum. Liquids transfer heat far more efficiently than air, which allows systems to maintain stable temperatures even under extreme computational loads.
This efficiency shift has material benefits. Reduced reliance on fans and chillers cuts energy use, while tighter thermal control helps extend server lifespan and improve reliability. From an
environmental, social and governance (ESG) perspective, liquid cooling also opens the door to heat recovery, transforming waste heat from a liability into a potential asset.
The study ‘Data center waste heat for district heating networks: A review,’ published in the Renewable and Sustainable Energy Reviews journal suggests that recovered data centre heat
could meet up to ten per cent of Europe’s heating demand by 2030, if effectively captured and reused. Nordic countries are already demonstrating what is possible, with large-scale projects supplying district heating networks using recovered data centre heat.
The opportunity of heat reuse Reusing waste heat is not a silver bullet, but it represents a meaningful step toward balancing digital growth with climate responsibility. Indeed, the WEF’s Artificial
Intelligence’s Energy Paradox report notes that redirecting data centre heat is a “crucial step” in managing AI’s environmental impact, even though it must sit alongside broader efficiency
measures.
However, heat reuse only works if thermal systems are designed and maintained correctly. Cooling fluids must remain thermally stable, pumpable and uncontaminated to ensure consistent heat
transfer. Poor fluid condition undermines efficiency, accelerates equipment wear and increases the likelihood of unplanned shutdowns — outcomes no data centre can afford.
Reducing energy waste in data centres
Let’s consider the example of a mid-scale European data centre that supports cloud and AI workloads. The facility operates high-density racks with an increasing proportion of liquid-cooled infrastructure. While the data centre invests heavily in electrical resilience and server efficiency, thermal fluid condition is monitored only intermittently, with maintenance triggered reactively by alarms or declining performance.
As processing demand intensifies, heat loads rise and cooling efficiency begins to drift. Pump energy consumption increases, PUE trends upward, and engineers identify early signs of fluid
degradation during targeted testing. This scenario reflects a wider industry challenge: cooling already accounts for 30–40 per cent of total data centre electricity use, leaving little room for
unnoticed losses.
To address this, the facility could implement a proactive fluid management strategy centred on routine sampling, laboratory analysis and trend-based diagnostics to identify early signs of degradation, contamination or inhibitor depletion before performance is affected. Pairing this approach with carefully specified thermal fluids supports stable heat transfer and consistent flow characteristics across high-density cooling loops.
Selecting fluids based on properties including thermal conductivity, freeze protection and material compatibility helps optimise pump efficiency, stabilise PUE and reduce the likelihood of thermal
stress or unplanned outages. When integrated with heat recovery, for example by redirecting higher-grade waste heat to nearby low-temperature heating networks, this approach can further improve overall energy utilisation while strengthening operational resilience and supporting ESG targets.
Designing for resilience, not reaction
While illustrative, this example reflects a reality many operators now face. Data centres are no longer static facilities — they are evolving, high-intensity environments where thermal performance directly influences cost, carbon footprint and continuity of service.
Predictive, data-led maintenance is increasingly recognised as a core operational discipline. In fact, predictive maintenance is one of the most effective ways digital technologies can reduce
energy waste across the value chain and thermal fluid management is no exception.
As global data creation accelerates — forecast to approach 400 zettabytes (ZB) by 2028, according to industry estimates — the margin for inefficiency continues to shrink. Cooling strategy can no longer sit quietly in the background.
For data centre operators, the message is clear: proactive cooling and fluid management turn a hidden operational risk into a controllable, strategic asset. In an industry defined by uptime, trust and sustainability, that shift may prove just as critical as the next generation of servers themselves.
To learn more about thermal fluid management solutions for data centres, visit Global Heat Transfer’s website at: http://www.globalhtf.com