Analyst(s): Nick Patience
Publication Date: January 22, 2025
GPU Cloud provider CoreWeave has operationalized two state-of-the-art data centers in Crawley and London Docklands, UK, featuring NVIDIA H200 GPUs and Quantum-2 InfiniBand networking. As the UK embraces its AI Opportunities Action Plan, these data centers have positioned CoreWeave as a key player in supporting the nation’s AI growth strategy.
What is Covered in this Article:
- CoreWeave’s launch of UK-based data centers
- How data centers facilitate advances in AI
- Role of NVIDIA H200 GPUs in driving AI innovation
- Strategic alignment with the UK’s AI Opportunities Action Plan
The News: CoreWeave launched its first UK-based data centers in Crawley and London Docklands. These facilities are equipped with NVIDIA H200 GPUs and Quantum-2 InfiniBand networking. They are designed to support large-scale AI workloads. Initially a part of CoreWeave’s £1 billion investment in the UK, this commitment expanded to £1.75 billion in October 2024, further bolstering the company’s AI infrastructure footprint.
CoreWeave Stakes Claim in UK AI Market with Advanced GPU Infrastructure Build-Out
Analyst Take: The rapid evolution of AI technology is reshaping the data center market, driving strong demand for advanced infrastructure. Hyperscalers, which essentially operate as large-scale data centers to support AI and cloud services, are at the core of this transformation. CoreWeave’s partnerships with Digital Realty for the Crawley center (live since October 2024) and Global Switch for the London Docklands center (operational from December 2024) reflect the company’s aggressive ambitions when it comes to AI infrastructure. Featuring NVIDIA H200 GPUs, these data centers mark a milestone in CoreWeave’s European expansion and align with the UK government’s AI Opportunities Action Plan.
Why Data Centers Matter in the AI Era
Data centers specialized in GPU computing provide the computational infrastructure necessary for generative AI and machine learning workloads, requiring specific architecture for power delivery, cooling systems, and interconnect technologies. These facilities differ from traditional data centers through their high-density GPU clusters, optimized networking fabrics, and specialized power management systems designed for parallel processing. The infrastructure includes technical requirements such as low-latency connections, custom hardware configurations, and scalable resource management. According to McKinsey, global data center capacity demand is projected to grow by 19%-22% annually between 2023 and 2030, driven by increasing AI adoption. CoreWeave operates as a specialized cloud infrastructure provider, focusing on GPU-accelerated computing with facilities designed specifically for high-performance computing workloads, distinct from traditional cloud providers who serve more general-purpose computing needs, though AWS and Google Cloud also have their own silicon optimized for AI and Microsoft is starting to roll its own out but is behind AWS and Google in this regard.
NVIDIA H200 GPUs: A Cutting-Edge Solution?
Although CoreWeave’s UK data centers are not equipped with NVIDIA’s latest Blackwell chips, which were released in late 2024, they utilize the advanced capabilities of the Hopper-generation H200 GPUs, introduced in 2023. These GPUs offer up to 141 GB of HBM3e memory, delivering a remarkable memory bandwidth of 4.8 TB/s and nearly 4 petaFLOPS of sparse FP8 performance. Such specifications make the H200 GPUs ideal for resource-intensive AI tasks, including training and inference of next-generation AI models. For instance, the H200 can fully run Meta’s Llama 3.1 405B model at 16-bit resolution on a single system – a task that previously required splitting workloads across multiple nodes with the older H100. While not NVIDIA’s newest offering, the H200’s exceptional performance, memory capacity, and efficiency position it as a powerful component of CoreWeave’s AI hyperscaling efforts.
CoreWeave’s Role in the UK’s AI Ambitions
The establishment of CoreWeave’s data centers in the UK is not just a mere expansion of the company’s business footprint but a strategic alignment with the country’s long-term AI aspirations. The UK government’s AI Opportunities Action Plan announced on January 13, 2025 incorporates 50 key recommendations from technology entrepreneur Matt Clifford to solidify the UK’s status as a global AI hub. These initiatives include the creation of AI Growth Zones and a significant expansion of sovereign compute capacity to accelerate the development of data centers while fostering private-sector partnerships. CoreWeave’s investment aligns with these initiatives, allowing the company to scale its operations in a supportive regulatory environment. In addition, newly adopted measures such as improved access to public data and targeted regulatory enhancements should enable CoreWeave to leverage the UK’s planned twentyfold increase in public compute capacity.
What to Watch:
- Competitive pressures will rise as rivals such as AWS, Microsoft Azure, and Google Cloud scale their AI-focused data centers and increase their levels of capex spending. CoreWeave’s ability to set itself apart through GPU-optimized infrastructure will become harder in the next few years.
- As AI workloads become more complex, the adaptability of CoreWeave’s infrastructure to support next-gen models, including those leveraging Blackwell GPUs, will become a focal point.
- Hyperscalers prioritizing renewable energy in operations could redefine market standards. CoreWeave’s efforts to integrate sustainability into its facilities will be critical to long-term growth and positioning.
- The UK government’s AI Opportunities Action Plan could accelerate public-private partnerships and infrastructure support. CoreWeave’s alignment with these initiatives may shape its competitive advantage.
Disclosure: The Futurum Group is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.
Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of The Futurum Group as a whole.
Other insights from The Futurum Group:
Will Google’s AI Enhancements Help Drive Greater User Adoption?
Redefining AI, Hybrid Cloud, and Enterprise Workloads with Cooling Innovations
CoreWeave Selects Dell Servers to Power GPU Cloud
Author Information
Nick Patience is VP and Practice Lead for AI Platforms at The Futurum Group. Nick is a thought leader on AI development, deployment, and adoption - an area he has researched for 25 years. Before Futurum, Nick was a Managing Analyst with S&P Global Market Intelligence, responsible for 451 Research’s coverage of Data, AI, Analytics, Information Security, and Risk. Nick became part of S&P Global through its 2019 acquisition of 451 Research, a pioneering analyst firm that Nick co-founded in 1999. He is a sought-after speaker and advisor, known for his expertise in the drivers of AI adoption, industry use cases, and the infrastructure behind its development and deployment. Nick also spent three years as a product marketing lead at Recommind (now part of OpenText), a machine learning-driven eDiscovery software company. Nick is based in London.
