• • • DATA CENTRES • • • The right decision depends on what the site
can tolerate and what the operator is trying to optimise: energy efficiency, water use, CAPEX, available space, maintenance access, resilience or long-term operating cost. Cooling must be treated as a whole-life
infrastructure decision, balancing current thermal requirements with future rack densities, local climate, water availability, energy costs, CAPEX, maintenance requirements and the consequences of downtime.
Efficiency economics The true economic impact of cooling has to be measured across the lifetime of the data centre, taking account of energy consumption, maintenance, replacement cycles, available floor space and the computing capacity the system enables. It cannot be assessed on capital cost alone. Using less water on-site does not always mean
lower overall costs or environmental impact, because the system may require more energy or higher upfront investment. A dry system may reduce on-site water
consumption but require more electrical energy; a wet or hybrid system will reduce energy demand at peak temperatures but consume more water and may add potentially costly treatment, controls and maintenance requirements. Liquid and immersion systems may enable more compute per square metre, but their CAPEX, integration and service requirements must be assessed alongside the value of the capacity they unlock. To help with this, Güntner has developed
IMPACT LCI, its life-cycle comparison tool integrated into its Product Calculator. IMPACT LCI enables users to compare potential energy consumption, operating costs and CO emissions from the planning stage onwards, helping move the discussion beyond initial capital cost towards the total value and environmental impact of a cooling solution throughout its operating life. To put this into perspective, the International
Energy Agency estimates that servers account for around 60 per cent of electricity use in modern data centres, while cooling can represent approximately 7 per cent of consumption in efficient hyperscale facilities, rising to over 30 per cent in less-efficient enterprise sites. This makes the choice of cooling technology a
significant operational decision: a lower upfront investment may result in higher energy and maintenance costs over time, while a more efficient and scalable system can help operators protect margins, maximise usable capacity and reduce the risk of expensive retrofits as workloads become more demanding.
Operational resilience Another important factor to consider is the operational resilience of the cooling system. With the recent heatwaves and associated strains on infrastructure it is vital that cooling continues to perform during periods of extreme weather, and can cope with planned maintenance, failures and disruption to power. Data centres underpin public services, businesses and wider economic activity, making continuity a strategic requirement, not simply a nice to have.
electricalengineeringmagazine.co.uk ELECTRICAL ENGINEERING • SEPTEMBER 2026 29 Resilience needs to be designed against the
conditions that create the greatest risk, including extreme temperatures, sudden load increases, loss of a cooling unit, planned maintenance and disruption to power or water supply. Cooling design should consider appropriate redundancy alongside independent circuits,
isolation capability, standby capacity, real-time monitoring, condition-based maintenance and the ability to service equipment without interrupting critical workloads. As rack densities increase, resilience also
depends on whether the heat-rejection system can scale without creating a new single point of failure. These decisions are not just important for the
initial use and the short-term, but must also take into account the future demands that could be placed on the cooling systems.
Future proofing Future-proofing a data centre means designing for the demands that are likely to arrive, not only those that exist at commissioning. Practical measures can include modular heat-
rejection capacity, reserved space and connection points, liquid-cooling-ready distribution, accessible maintenance routes, scalable controls and clear provisions for additional pumps, heat exchangers or cooling modules. Modular, liquid-cooling-ready or hybrid systems can provide a practical route forward, allowing
capacity to grow in line with demand while helping protect the data centres operating life and long- term efficiency.
Sustainability As well as predicting future growth and operational security, sustainability must also come into the mix. For Güntner the issue of sustainability is
embedded into the very fabric of the business, through its pledges Güntner has clearly stated its sustainability targets. These include reducing carbon intensity across its manufacturing facilities by 40 per cent, reducing energy intensity by 10 per cent, and reducing water intensity by 30 per cent by 2030. For data centre operators, these commitments
should be music to their ears. These pledges align closely with the priorities of data centre operators, and should help them lower operational energy use, reduce water demand, and provide more responsible refrigerant choices and equipment designed with its full lifecycle in mind. As AI workloads become more demanding,
cooling for data centres can no longer be treated as a supporting function. It is perhaps the most important strategic infrastructure decision that influences operating cost, usable capacity, sustainability performance and resilience over the facility’s lifetime.
www.guntner.com
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48
