New data-centre partnership gives Microsoft access to Nvidia chips

Microsoft has reached a $9.7 billion agreement with data-centre operator IREN, gaining access to advanced Nvidia chips that could ease the strain on the company’s infrastructure and support AI services such as ChatGPT. Limited access to computing power has slowed growth across the sector, and large tech firms have pointed to capacity shortages during recent earnings reports.

The impact of the deal was felt on the market. IREN shares climbed as much as 24.7% to a record on Monday before easing to a smaller gain. Dell, which will supply IREN with Nvidia’s GB300 chips and related equipment, was also trading slightly higher. Microsoft plans to use about $5.8 billion worth of that hardware.

The five-year agreement gives Microsoft a way to increase its computing capacity without building new facilities or sourcing more power. Those two challenges have made it harder for the company to keep up with rising demand for AI services. The arrangement may also help Microsoft avoid heavy spending on hardware that could become outdated as new processors enter the market.

Companies known as “neocloud” providers, including CoreWeave and Nebius Group, have grown quickly by offering cloud services built on Nvidia chips. Microsoft recently signed a $17.4 billion deal with Nebius for more infrastructure access. These partnerships reflect the industry’s effort to secure hardware as interest in AI grows.

IREN has multiple data centres across North America with a total capacity of 2,910 megawatts. The company’s market value stood at $16.52 billion as of its previous close, after its shares rose more than sixfold this year. IREN said the Nvidia processors will roll out in stages through 2026 at its 750-megawatt campus in Childress, Texas. The company is also planning liquid-cooled data centres that could support about 200 megawatts of IT capacity.

Microsoft’s prepayment will help cover part of IREN’s $5.8 billion agreement with Dell. The contract includes a clause that allows Microsoft to walk away if delivery goals are not met.

A separate deal was also announced on Monday. AI cloud startup Lambda revealed a multibillion-dollar agreement with Microsoft to deploy AI infrastructure using Nvidia hardware. While the exact value was not shared, the company said the plan would include tens of thousands of Nvidia GPUs, including GB300 NVL72 systems, which were first introduced earlier this year.

“It’s great to watch the Microsoft and Lambda teams working together to deploy these massive AI supercomputers,” said Stephen Balaban, CEO of Lambda, in a statement. “We’ve been working with Microsoft for more than eight years, and this is a phenomenal next step in our relationship.”

Microsoft launched its first Nvidia GB300 NVL72 cluster in October, giving the company more options to scale AI workloads beyond its traditional data-centre footprint. These deals signal that competition for computing power may continue as AI adoption spreads across the industry.

(Photo by Simon Ray)

See also: UK watchdog flags Microsoft and Amazon for stifling cloud competition

Want to learn more about Cloud Computing from industry leaders? Check out Cyber Security & Cloud Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events, click here for more information.

CloudTech News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here.