<img height="1" width="1" style="display:none;" alt="" src="https://px.ads.linkedin.com/collect/?pid=1448210&amp;fmt=gif">
Free Research

Hyperscaler Data Center Capacity to Surge More Than 6x by 2035 as AI and Cloud Expansion Reshape Global Infrastructure

New ABI Research market data finds hyperscalers are scaling power far faster than site counts, signaling a decade defined by larger campuses, denser compute, and AI-ready capacity planning

05 May 2026

Active hyperscaler IT load worldwide will rise from 24.37 Gigawatts in 2025 to 147.13 Gigawatts by 2035, according to global technology intelligence firm ABI Research. The new forecast shows hyperscaler infrastructure entering a new phase of expansion in which power, cooling, and workload density are scaling much faster than physical site counts, reflecting the combined impact of AI build-outs, core cloud growth, storage demand, and enterprise migration workloads. ‌

“The hyperscaler market is no longer defined simply by how many facilities operators build, but by how much compute capacity they can concentrate into large, power-dense campuses,” said Leo Gergs, Principal Analyst at ABI Research. “AI is a major accelerator, but the broader story is that hyperscalers are simultaneously supporting next-generation AI clusters and the continued expansion of mainstream cloud and enterprise workloads.”

ABI Research’s data indicates that the total number of hyperscaler sites will increase only modestly, from 3,182 in 2025 to 3,558 in 2035, while facilities above 10 Megawatts will account for nearly all meaningful growth, rising from 2,921 to 3,255 sites. Over the same period, available capacity is projected to expand from 33.96 Gigawatts to 228.96 Gigawatts, while under-the-roof capacity will grow from 45.08 Gigawatts to 276.63 Gigawatts, underscoring the industry’s move toward deep, multi-phase build-ahead strategies. ‌

At the same time, energy generation and grid constraints are emerging as critical bottlenecks for hyperscaler expansion. As campuses scale into multi-hundred-megawatt deployments, operators are prioritizing access to high-capacity transmission, long-term power agreements, and dedicated energy sources. In many regions, limited grid availability, interconnection delays, and policy constraints are making power access—not land or capital—the primary gating factor for growth.

The competitive landscape is led by AWS, Microsoft Azure, and Google Cloud, alongside players such as Meta, Oracle, major U.S. hyperscale-colocation providers, and Asia-Pacific leaders including Alibaba Cloud, Tencent Cloud, and Huawei Cloud. Regionally, North America leads, Asia-Pacific is growing fastest, Europe is consolidating into fewer high-capacity sites, and the Middle East and Africa are advancing national-scale cloud campuses.

“Over the next decade, hyperscaler success will depend on the ability to secure grid-adjacent capacity, deploy advanced cooling architectures, and align infrastructure planning with changing workload composition,” Gergs said. “The market opportunity will favor operators that can translate power access and capacity readiness into scalable AI and cloud platforms without compromising long-term efficiency or regional compliance requirements.”

These findings are from ABI Research’s Hyperscaler Data Centers Market Data Overview: 2Q 2026 report, part of the company’s Cloud research service, which includes research, data, and ABI Insights. The dataset introduces full capacity-state modeling across active, available, and under-the-roof capacity, along with a redesigned workload split between AI and legacy compute to support cross-segment data center analysis. ‌

applicationoctet-stream-May-04-2026-01-02-35-2812-PM

 

Contact ABI Research

Media Contacts

Americas: +1.516.624.2542
Europe: +44.(0).203.326.0142
Asia: +65 6950.5670

Related Research

Related Service