Has x86 lost the data center battle? Arm claims victory as it declares close to 50 percent of compute shipped to top hyperscalers in 2025 will be Arm-based

• Arm says Neoverse powers nearly half of 2025 hyperscaler compute shipments
• AI growth makes power-efficiency essential for hyperscale data centers
• AWS, Google, Microsoft are leading shift from x86 to Arm infrastructure

Arm says its Neoverse platform is becoming the architecture of choice for the cloud, as the likes of AWS, Google Cloud, Microsoft Azure and others drive a broader shift away from x86 in the data center.

Mohamed Awad, SVP and GM of the Infrastructure Business at Arm, says power efficiency and scalability are redefining infrastructure at hyperscale. “Just over six years ago, we launched Arm Neoverse for the next-generation of cloud infrastructure, recognizing a world where delivering new levels of scalable performance on top of Arm’s flexible and power-efficient compute platform could enable a systemic shift in the capabilities and costs of the data center ecosystem,” he explained.

“Fast forward to today, the adoption of Neoverse has reached new heights: close to 50 percent of the compute shipped to top hyperscalers in 2025 will be Arm-based,” he added. Much of this momentum can, inevitably, be attributed to the rise of AI.

AI servers set to grow by 300 percent

AI is reshaping cloud infrastructure, driving explosive growth in compute demand and forcing hyperscalers to prioritize power efficiency at massive scale. Data centers are being designed in gigawatts, not megawatts, making efficiency a requirement rather than a differentiator, something that has been central to Arm's architecture for decades.

That's not the only reason for the architecture's success of course.

"The Arm compute platform is additionally giving our partners the flexibility to create a new generation of customized, differentiated silicon solutions for AI. For example, Nvidia’s Grace Blackwell superchip for AI-based infrastructure combines Nvidia’s Blackwell GPU accelerated computing architecture with the Arm Neoverse-based Grace CPU, integrated with an extraordinarily high bandwidth, coherent mesh network – a system tailor-made to achieve unmatched performance for AI workloads," Awad says.

With ten of the world’s largest hyperscalers developing and deploying Arm-based chips, the future looks bright for the company - but that dynamic could shift if, as rumored, Arm decides to produce its own silicon.

Speculation on that front gained signifcant momentum after Arm’s Japanese owner, SoftBank, acquired Arm-based chipmaker Ampere for $6.5 billion, raising concerns that Arm could end up competing directly with its partners in the fast-growing data center market.

You might also like

• Softbank buying Ampere for $6.5 billion and could integrate it with Arm
• Is an AMD Arm superchip in the works?
• Arm wants to build its own CPU server for hyperscalers, reports say