AMD and Rackspace Technology signed a definitive agreement to deploy an initial 30 MW footprint of AMD-based compute across Rackspace’s global data centers. The rollout is scheduled to begin in late 2026 through 2028 and follows a Memorandum of Understanding announced May 7, 2026.
The deployment positions AMD as a strategic technology partner at the silicon layer of Rackspace’s governed AI stack. Rackspace plans to use the dedicated capacity for regulated enterprise workloads, including healthcare providers interested in accelerated compute for clinical AI and inference at scale.
The collaboration will combine AMD Instinct GPUs, including MI355X, MI350P, and future successor solutions, with AMD EPYC CPUs inside Rackspace’s Enterprise AI Cloud architecture. The companies said the integrated approach will allow Rackspace to direct workloads to appropriate compute resources while maintaining accountability for performance and outcomes.
