Trial / Demo
Gary O’Keeffe is the Chief Growth Officer for Ethos Engineering – Ireland's largest Mechanical Electrical consultancy and one of Europe's leading experts in Data Centre design, Sustainability, Urbanism, Global Tech, and Digital Services (Smart Buildings). He leads a multi-disciplinary engineering team, specialising in data centre project delivery, from concept to project handover – and it’s this day-to-day involvement that gives him a unique perspective in the industry.
In this blog, he shares his expertise on the challenges he faces and how early stage modelling truly impacts his teams. These insights are also featured in the IES whitepaper: De-risking High-Performance Data Centres with Dynamic Simulation. As data centres face mounting pressure to support increasingly power-hungry AI and GPU workloads, the stakes for getting design and retrofit decisions right have never been higher. Gary brings a ground-level perspective on how engineering teams are navigating this complexity – balancing speed, resilience, and operational continuity in parallel.
Below, he outlines the key challenges shaping modern data centre delivery and why the tools teams use at the earliest stages of a project can make all the difference.
Going offline isn’t an option
I think that retrofit and expansion projects face some of the sector’s tightest technical and commercial constraints. And when you consider that legacy infrastructure, partial loads and evolving cooling strategies must be reconciled – all while keeping facilities online.
To support this, I truly see the growing value in early stage models that consolidate airflow behaviour, load changes, temperature profiles and equipment interaction into a single environment. It means our teams regularly receive quick direction to prevent programme drift. In fact, we often sit in a room and say, “It would be great to tell them this tomorrow,” which just underscores the need for tools that deliver quick, credible insight.
One model. One goal.
I also think that bringing these factors under one model could be particularly transformative in retrofit scenarios, where constraints are often interdependent. It can only lead to faster approximation of cooling feasibility, resilience under heat events and potential design limits to enable my teams to eliminate non-viable options early and focus investment where it delivers genuine value.
Get even more insight
Gary’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.
Visit our website to discover how whole-facility modelling is transforming data centre planning, design, retrofit, and operation in an era of AI-driven infrastructure demands
