AMD and Rackspace Technology, an enterprise AI infrastructure and solutions company, have agreed to gradually deploy AMD-based computing infrastructure in Rackspace's global data centers. The two companies plan to jointly establish governance-based enterprise AI infrastructure targeting regulated industries and environments with high demands for data sovereignty.

AMD and Rackspace announced on the 17th that they have signed a contract to build an initial 30MW dedicated computing system.

Construction will proceed from the end of 2026 to 2028. This agreement fulfills the Memorandum of Understanding (MOU) announced on May 7, in which AMD participates as a technology partner responsible for the core computing layer in Rackspace's governance-based AI stack.

Once the full deployment is complete, 30MW of AMD-dedicated computing resources will be deployed across the Rackspace infrastructure.

Both companies explained that this enables them to support large-scale enterprise AI workloads in regulated industries, including healthcare.

It is reported that this will particularly address the demand from healthcare providers interested in accelerated computing for clinical AI and large-scale inference workloads.

The architecture integrates AMD Instinct GPUs MI355X and MI350P, next-generation successors, and AMD EPYC CPUs.

"Companies in regulated industries need AI infrastructure with governance built in from the start, and must have a system where a single operating entity takes responsibility for business performance," said Rackspace Technology CEO Gajen Kandia.

Dan McNamara, Senior Vice President of Compute & Enterprise AI at AMD, stated, "By combining AMD's AI computing solutions with Rackspace's governance-based cloud operating model, we support the construction of AI infrastructure that is open, scalable, and accountable."

Both companies will invest sales and marketing resources to identify enterprise customers requiring AMD computing-based infrastructure and deploy dedicated personnel to jointly develop customer opportunities across regulated industries.

This contract is expected to accelerate the provision of the four integrated capabilities presented in the MOU: Enterprise AI Cloud, Enterprise Inference Engine, Inference as a Service, and Bare Metal AMD Instinct.

Both companies fully manage everything from bare metal computingThey added that they aim to build a managed enterprise AI infrastructure market that encompasses even type inference services.