Skip to content
Type
Explainer

Understanding data centres: how engineers can power change

Date
24 November 2025

Experts came together to discuss the opportunities and challenges presented by this vital – yet often misunderstood – infrastructure.

Understanding data centres: how engineers can power change
About 80% of the UK's data centre facilities can be found in the Greater London area. Image credit: Shutterstock

When you stream a video, send an email, or use any app on your phone, you’re relying on infrastructure most people never see.

Data centres are the backbone of our digital lives, yet they are often misunderstood, reduced to headlines about power consumption.

It’s time to look beyond the noise and understand what data centres are and why they’re essential to our future.

What is a data centre?

At its core, a data centre is a facility that houses the servers, storage systems, and networking equipment that keep our digital world running.

Every time you watch Netflix, work on a cloud document, or check your banking app, you’re connecting to a data centre somewhere.

They’re the invisible infrastructure making modern life possible, from healthcare systems to financial services, from education platforms to emergency response networks.

The UK government's recent designation of data centres as critical national infrastructure recognises this reality.

The value we often overlook

The conversation around data centres often focuses on what they consume. But what about what they enable?

“Challenges around data centres are exacerbated at the moment,” said Luisa Cardani, head of data centres programme at techUK. She was speaking at an event at the ICE’s headquarters, titled Engineering London’s Data Centres – Navigating Planning, Sustainability and Grid Resilience.

“If you want to watch YouTube, Netflix or use an online banking app, or NHS digital services, people need to know that you need infrastructure for this to happen,” Luisa said.

It’s a simple statement that reminds us that the digital services we rely on don’t just exist in the cloud. They exist in physical buildings with real infrastructure needs.

For engineers and planners, data centres represent both opportunity and challenge.

Changing the narrative

One of the biggest barriers to progress is not technical, it’s perception.

“Data centres are often viewed as power-hungry, but their broader value to society is frequently overlooked,” said Dr Anna Dowson, senior director at GreenScale.

This perception problem matters because it affects planning decisions, community support, and policy development.

The reality is more nuanced.

Increasing efficiency

Yes, data centres are energy-intensive, but they are also becoming more efficient.

The industry is innovating rapidly, exploring air cooling systems, optimising designs, and pushing for smaller footprints.

Technology becomes more efficient with use, and data centres are no exception.

Communicate the benefits

But innovation alone will not solve the perception problem.

As Chhavi Nayak, senior manager at Deloitte Capital Projects, said: “We need to communicate the value of data centres to the public and focus on effective ecosystem orchestration.”

“This is not about PR spin. It is about honest conversation about trade-offs, benefits, and the infrastructure choices we face as a society”.

Dries Hagen, client lead and senior project manager at Bryden Wood, takes this further: “If technology is to become life enhancing, then data centres and the technology within need to come out of the shadows and be more mainstream in the psyche of the communities which they serve.”

The challenges we cannot ignore

80% of the UK's data centre facilities can be found in the Greater London area. This clustering creates real tensions.

West London is facing electricity capacity shortages that could delay housing delivery for the next decade. This directly conflicts with the government's target of building 1.5 million new homes over five years.

Javier Zamorano, senior director at Laminar, offers his perspective: “We focus on London due to the great connectivity and the need for low latency [see definition below].

“However, we must remember that the UK has a high population density, so there’s need for low latency in many other regions which are also very well connected.”

Building together

The path forward requires collaboration.

Government, industry, planners, and communities need to work together on frameworks that balance competing demands.

For civil engineers and students entering the field, data centres represent a frontier.

They are projects where sustainability challenges meet cutting-edge technology, where infrastructure constraints require creative solutions, and where your work directly shapes how society functions.

What comes next

The digital economy is not slowing down.

AI, cloud computing, and emerging technologies will only increase demand for data centre capacity.

The question is not whether we need this infrastructure, it is how we deliver it sustainably, equitably, and intelligently.

That requires engineers who understand technical excellence and community impact.

It requires planners who can balance competing priorities. And it requires all of us to engage with these questions before they become crises.

Data centre terms made easy

We have put together a list of commonly used data centre words paired with simple explanations so you can navigate the terminology with confidence.

  • Latency: the delay (in milliseconds) it takes for data to travel from one point to another.
  • AI training data centre: a type of AI-focused data centre built to ‘train’ AI models using very large datasets.
  • Inference data centre: a data centre optimised for running inference, which is when trained AI models are used to make predictions or decisions in real time.
  • Hyperscale data centre: a large data centre built to support thousands of servers and massive workloads, typically used by big tech companies (hyperscalers).
  • Edge data centre: a smaller facility closer to end-users that reduces latency by keeping data nearer to where it is needed.
  • Colocation: a service where customers rent space in a data centre to house their servers instead of keeping them on their own premises.
  • Cloud: servers accessed over the internet, hosted in data centres, instead of being kept in your own office/premise.
  • Redundancy (N, N+1, 2N): extra equipment installed so there is a backup if something fails.
  • PUE (Power usage effectiveness): a measure of how energy-efficient a data centre is.
    • 1.0 = all energy goes to IT/server equipment (an idealised goal)
    • Higher than 1.0 = extra energy used for cooling, lights, etc. (current reality)
    • Current UK PUE average figures running at between c. 1.5-1.8. Best in class would be 1.2
  • SLA (service level agreement): contractualised agreement between a data centre developer/operator and the tenant/occupant, that guarantees certain levels of service, like uptime or response times.
  • Rack: a metal frame that holds servers and other IT equipment in neat rows. Think of it like a bookcase for computers.
  • Server: a powerful computer that stores, processes, and delivers data to other computers.
  • White space: the main area of the data centre where IT equipment like servers and racks are installed.

  • Roxana Hurjui, communications lead – London, South East & East of England at Institution of Civil Engineers