Reliability

One Substation

Late on Thursday 20 March, a fire broke out at the North Hyde electrical substation in west London, one of three supply points feeding Heathrow Airport. Heathrow closed for most of Friday 21 March. More than 1,300 flights were cancelled and around 270,000 passenger journeys were disrupted. On Saturday the Energy Secretary asked the National Energy System Operator (NESO) to lead a review.

The question everyone asked: how can the UK’s busiest airport be taken out by one fire, when it has three supplies?

Redundancy Isn’t the Same as Failover

We’ll learn the details from the review. But the pattern is one I see in IT all the time:

  • Having a second supply (or a second data centre, or a second region) is the easy part.
  • Moving the load onto it — safely, quickly and in the right order — is the hard part.
  • Large sites often can’t switch over instantly. Internal networks have to be reconfigured, systems restarted in sequence, and safety checks done before people can come back in.

In data-centre terms: having a DR site doesn’t mean you have a DR capability. What matters is how long the switchover takes, and whether anyone has practised it at full scale.

What I’d Ask of Any Critical Service

  1. If supply A disappears right now, how long until we’re running on B, and is that time written down and tested?
  2. Which systems need to be brought up in order, and who holds that sequence?
  3. Do our upstream providers know that we are critical to them? Do they know what we’ll do when they fail?

I’ll come back to this when the review is published.

More Information

Postscript: NESO’s final report in July 2025 found the fire was caused by a known fault, a moisture problem first identified in 2018 but never fixed. It also found the design of Heathrow’s private internal power network limited how quickly the airport could switch to its other supply points.