Friday, June 26, 2026
EN·DarkSubscribe
AI Infrastructure · News & Analysis
HomeCompute & CloudReport
Compute & Cloud · Report

CoreWeave, a leading GPU cloud provider, has opened two H200 GPU-equipped datacenters in the United Kingdom.

Expands CoreWeave's geographic footprint and capacity, consolidating its position as a top-tier GPU cloud peer.
Trade pressSlicast · January 13, 2025 · Global · Source: theregister.com
importance 75

CoreWeave, a New Jersey-based datacenter operator, has announced two new GPU-packed facilities in the United Kingdom as the country reaffirms its ambitions to become an AI superpower. Located in Crawley and the London Docklands and developed in collaboration with colocation providers Digital Realty and Global Switch respectively, the two sites are part of a £1 billion ($1.2 billion USD) investment by CoreWeave. "The UK is an important market for CoreWeave, with our European headquarters here and further operational plans for the country," said Mike Mattacola, chief business officer at CoreWeave. "We are pleased to be partnering with Digital Realty and Global Switch to deliver the next generation of AI infrastructure in the UK."

The datacenters are equipped with Nvidia's H200 accelerators and 400 Gb/s Quantum-2 InfiniBand networking, based on the Hopper architecture rather than the newer Blackwell chips that began making their way to market in Q4 2024. Introduced in 2023, the H200 is among Nvidia's most advanced GPUs of the Hopper generation, featuring up to 141 GB of speedy HBM3e memory, 4.8 TB/s of memory bandwidth, and nearly 4 petaFLOPS of sparse FP8 performance per GPU.

With more than a terabyte of high bandwidth memory across eight GPUs per server, CoreWeave's UK datacenters are equipped to support frontier models that have stretched existing systems to their limits. A single H200 system can now run Meta's largest model Llama 3.1 405B at full 16-bit resolution, whereas previously the model would have had to be distributed across two nodes or quantized to 8-bits to run on an H100-based system. Past deployments have typically involved clusters of 10,000 or more accelerators, though the exact number of GPUs at the Crawley and London Docklands sites has not been disclosed.

These UK datacenters represent the latest expansion in CoreWeave's rapid global growth. The company announced London as the location of its European headquarters last spring with an initial investment of £1 billion, which was subsequently inflated to £1.75 billion ($2.1 billion USD) in October during the International Investment Summit. CoreWeave currently operates 28 datacenters around the globe with plans to bring 10 additional facilities online by the end of 2025.

Meanwhile, the British government is doubling down on its efforts to establish itself as an AI superpower, having outlined the AI Opportunities Action Plan on Monday with 50 recommendations aimed at boosting productivity by 1.5 percent a year, with an expected economic impact of £47 billion ($57 billion) annually. However, supporting the additional AI infrastructure necessary to realize these plans may prove troublesome for the UK's National Grid, as its CEO has previously warned.

Read the original
CoreWeave, a leading GPU cloud provider, has… · Slicast