AMD and Rackspace Technology have signed a definitive agreement to deploy an initial 30 megawatts (MW) of AMD-powered AI compute infrastructure, marking a strategic push into governed enterprise AI systems designed for regulated industries such as healthcare and finance. The rollout will begin in late 2026 and continue through 2028 across Rackspace’s global data center footprint.
The collaboration formalizes an earlier memorandum of understanding and positions AMD as a core silicon partner within Rackspace’s enterprise AI stack, combining high-performance compute with managed cloud operations for large-scale AI deployment.
30MW AMD AI Compute Targets Regulated Enterprise Workloads
At full scale, the 30MW deployment will support AI workloads requiring both high performance and strict governance controls. The infrastructure is designed to serve regulated sectors where data security, compliance, and operational accountability are critical.
Early enterprise interest includes use cases in healthcare, particularly in clinical AI and inference at scale, where fast processing and secure infrastructure are essential.
AMD Instinct GPUs and EPYC CPUs Power Hybrid AI Stack
The deployment will integrate AMD’s compute portfolio, including AMD Instinct GPUs (MI355X, MI350P, and future generations) and AMD EPYC CPUs.
These components will be embedded in Rackspace Technology’s Enterprise AI Cloud architecture, enabling dynamic workload routing between accelerated and general-purpose compute systems.
This setup is designed to optimize performance based on workload requirements, particularly for AI inference and enterprise-scale model deployment, where efficient balancing of GPU-accelerated and CPU-driven tasks is critical for scalability and cost efficiency.
Rackspace Positions “Governed AI Stack” for Regulated Industries
Rackspace Technology is positioning the collaboration as a governed alternative to fragmented AI infrastructure models. The company emphasizes a single-operator approach where accountability spans from hardware to outcomes.
Key focus areas include Enterprise AI Cloud infrastructure, Enterprise Inference Engine, Inference-as-a-Service offerings, and bare-metal AMD Instinct deployments.
According to Rackspace Technology, this model is designed to address enterprise demand for controlled, auditable AI systems as organizations move from experimental AI adoption to production-grade deployment, ensuring clearer governance and operational visibility across the full AI stack.
AMD Strengthens Enterprise AI Compute Strategy
AMD is expanding its role in enterprise AI infrastructure as demand grows for scalable and open compute ecosystems. The partnership with Rackspace aligns with AMD’s strategy of positioning Instinct GPUs and EPYC processors as core components of enterprise AI deployments.
The companies will also coordinate sales and marketing efforts to identify opportunities across regulated industries, signaling a more integrated go-to-market strategy for enterprise AI solutions.
Enterprise AI Demand Drives Shift to Managed Infrastructure
The collaboration reflects a broader industry shift toward managed AI infrastructure and inference services, as enterprises move from isolated AI experiments to integrated workflows inside core business systems.
With increasing demand for governance, scalability, and performance transparency, the AMD–Rackspace partnership highlights how infrastructure providers are evolving to support production-ready AI at scale.
As deployments ramp up through 2028, the initiative positions both companies in the growing market for regulated, enterprise-grade AI compute solutions.
