Trial / Demo
The data centre has fundamentally changed. AI and high-density workloads are significantly increasing thermal and electrical loads, requiring more robust cooling and power strategies. These elevated demands increase the likelihood of hotspots, operational vulnerabilities, capital investment risk and supply constraints if potential issues are not identified and accounted for early in the design process.
The new standard
Some studies indicate that traditional cooling technologies can compromise up to 40% of the total energy demand in a data centre. While air cooling has been the standard approach for many years, the industry is increasingly exploring more advanced liquid and hybrid cooling technologies, including direct-to-chip, free/evaporative/immersion cooling, rear door heat exchangers, and aisle containment strategies which can dissipate heat much more directly and efficiently. According to the Uptime Institute, more than one in five data centres (22%) already use liquid cooling, whilst another 61% are considering it for their facilities. Recent research by Microsoft indicates that these solutions can reduce greenhouse gas emissions by as much as 15–21%, energy demand by 15–20%, and blue water consumption by 31–52% compared to traditional air cooling systems. However, these solutions do not come without risk.
There’s no one size fits all solution
Design teams need to conduct robust analyses in order to determine the most appropriate HVAC solution based on a data centre’s particular site and climate. For example, would a hybrid solution perform better in one location than another? And how do different cooling strategies compare when evaluated against real weather patterns and operational profiles? Whilst always ensuring that the design comfortably meets key temperature and humidity design conditions.
Identifying climate-appropriate cooling strategies early in the design process enables operators to make informed decisions that balance performance, efficiency, and capital investment. In a recent project IES delivered for a hyperscale client in North America, dynamic simulation was used to assess a range of cooling strategies for an existing site located in a cool, dry climate. In this instance, direct evaporative cooling (DEC) emerged as the optimum solution, demonstrating a 16% power usage effectiveness (PUE of 1.6) improvement over air-cooled chillers and a 95% reduction in water usage effectiveness (WUE of 0.1) over water cooled alternatives. Projects like this are testament to the importance of ensuring cooling strategies are validated through a holistic, climate-specific analysis to ensure solutions are optimised for efficiency, sustainability and long-term operational resilience.
To learn more about dynamic simulation and how it addresses the key challenges facing modern data centres, download the full IES whitepaper: De-risking High-Performance Data Centres with Dynamic Simulation.
Discover how whole-facility modelling is transforming data centre planning, design, retrofit, and operation in an era of AI-driven infrastructure demands.
