Following our recent keynote discussion at Tech Show London, one theme came through clearly: the conversation around edge is no longer theoretical. It is moving quickly into real-world decision making.
Organisations are now grappling with a new set of trade-offs. The traditional choices e.g. build or buy, cloud or on-prem are being reframed by newer tensions: hyperscaler scale versus sovereign control, centralisation versus resilience, performance versus governance. All of this is in pursuit of something increasingly complex: delivering low latency, while maintaining security, resilience and control over data.
At the same time, the direction of travel is becoming clearer. The edge computing market is growing at sustained double-digit rates, with forecasts suggesting it could nearly double over the next few years (IDC). Gartner predicted that by 2025, 75% of enterprise data would be created and processed outside traditional data centres or cloud, and we are now starting to see that shift materialise in practice.
- Concentration of cloud providers: Ofcom estimates that AWS and Microsoft account for 70–80% of the UK cloud market, highlighting the level of dependency on a small number of providers.
- Sovereignty and control: Increasingly shaped by geopolitical considerations, regulatory pressure and a growing focus on where data is processed and governed.
- AI driving distribution: Research from STL Partners highlights how AI workloads are pushing compute closer to where data is created, driving demand for more distributed infrastructure.
What is perhaps most interesting is what this means in practice. For the last decade, the industry has largely optimised for scale and centralisation. But we may now be reaching a point where that model begins to rebalance. A useful analogy is something as simple as electric windows in cars. For years, winding a window manually was entirely normal. Then gradually, almost without noticing, the standard shifted, and today, the idea of going back feels unthinkable. Distributed compute may be approaching a similar inflection point. The building blocks are now in place, and external pressures, from AI to sovereignty to resilience, are accelerating adoption.
- Where is data actually processed and stored at any given moment?
- Who has operational and legal control over that data?
- How do you apply consistent security and governance across a distributed estate, not just within a centralised environment?
This is where micro-edge becomes meaningful. It is not simply about reducing latency. It is about narrowing the gap between performance, sovereignty and operational control, enabling organisations to design infrastructure that reflects both technical and strategic requirements.
There are also broader signals reinforcing this shift. The UK government has committed £2 billion to expand national compute capacity twentyfold by 2030, alongside the development of AI Growth Zones and a Sovereign AI Unit. This reflects a growing recognition that compute i.e. where it sits, how it is controlled, and how it is scaled is becoming a matter of national importance.
For those exploring this space further, this recent analysis from SDxCentral looks at how distributed infrastructure, particularly when anchored in existing physical networks, could reshape how and where compute is deployed.
As the conversation continues to evolve, one thing is clear: edge is no longer just about performance. It is about control, resilience and trust. And that is where the next phase of innovation will be defined.