The UK's £100 billion data center boom is suffering from a major bottleneck that puts capacity and performance at risk

The UK's £100 billion data center boom is encountering a significant bottleneck that threatens capacity and performance, according to a survey of senior data center decision-makers. The issue stems from cabling constraints that prevent data centers from scaling and futureproofing to meet the rapid advancements in AI technology. Onnec, a data center infrastructure specialist, found that 77% of UK data center operators are encountering critical bottlenecks in sourcing cabling to support their AI workloads.

The AI rush is creating additional cabling problems that could take years to resolve, and 37% of operators are experiencing latency or bottleneck issues with AI training.

Moreover, 65% of data center operators reported increased AI capacity, with 83% stating that upgrades are needed almost immediately upon hardware deployment to meet new demands and capacity. This compounds the cabling bottleneck, with 33% unable to scale their data centers without significant cabling work. Insufficient upfront cabling planning has caused project delays and additional costs in some cases, with 44% of operators noting the issue.

Data center infrastructure specialist Matt Salter emphasized the importance of incorporating cabling into the design process before finalizing rack layouts and pathways. He stressed that while power and GPUs are prioritized for AI, the network connecting the hardware often lags behind, leading to communication issues that hinder AI performance.

Salter advised operators to ensure sufficient capacity, access, and flexibility to accommodate more than the initial hardware deployment, as an improperly designed network can lead to expensive work inside live facilities much sooner than expected.