Trial / Demo
Eoin Kane is an Operations Director with Ethos Engineering who brings vast experience leading teams on numerous complex projects with demanding simultaneous schedules – all from initial assessment through to final commissioning.
Ethos Engineering utilises a talented, dedicated team of mechanical and electrical engineers to build a reputation for delivering innovative solutions on a diverse range of data centre projects across Europe.
In this blog he shares his expertise on how the rapid rise in high-density compute loads is challenging established data centre design conventions, and what engineering teams must do to stay ahead. These insights are also featured in the IES whitepaper De-risking High-Performance Data Centres with Dynamic Simulation.
Across the EMEA region, the assumptions that have long underpinned data centre design are being tested by a new generation of high-density, AI-driven workloads that behave in ways traditional models were never built to anticipate. For Eoin and his team at Ethos Engineering, keeping pace means going beyond standard sizing exercises to interrogate how a facility will actually perform under the full range of conditions it may face.
Below, he outlines the design challenges now defining the sector and why scenario-based, environmentally-aware analysis has become essential to delivering resilient, future-ready facilities.
Resilience is key
Not only have I noticed intensifying compute demands are redefining traditional design boundaries across the EMEA region, but there’s also been a real shift towards higher-density racks, and the pace of change is accelerating. As density increases and operating profiles become less predictable, operators are looking for greater assurance that designs can maintain resilience across a wider range of conditions. That’s why Ethos Engineering places strong emphasis on scenario-based analysis, supported by CFD modelling where appropriate, to better reflect real-world behaviour. In fact, we need to run many scenario-based studies, including assessments of grid-outage scenarios, thermal step changes, and phased expansion sequences.
The outside matters too
For me, external environmental effects, such as chiller derating at high ambient temperatures and air recirculation/re-entrainment risk, are also becoming major considerations, particularly on thermally constrained sites. And annual energy and PUE are still commonly derived using bin-method annualisation models, often fed by CFD outputs and binned weather data to represent different operating states. I also think there’s scope to improve both efficiency and accuracy by linking performance calculations more directly to scenario-driven and time-dependent analysis, better reflecting the variability now emerging as the norm.
Get even more insight
Eoin’s expertise is just part of a comprehensive analysis of how advanced modelling approaches are enabling data centre operators to build facilities that are more resilient, efficient, and prepared for the demands of AI infrastructure.
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.
