Digital Realty Trust, Inc.

New York Stock Exchange
Back to all articles
Bullish +65

Digital Realty Announces Transactions to Drive Continued Platform Growth - GlobeNewswire

🚀 Digital Realty announces ServiceFabric Model Context Protocol (MCP), an open standard enabling programmable controls for Private AI environments across its global platform.

🌐 The solution extends Digital Realty's Foundation for AI strategy, covering over 800 data centers and third-party facilities to support secure, distributed enterprise AI deployments.

🤝 Partnerships with ePlus, Lenovo, and Dell are established on infrastructure powered by NVIDIA and AMD technologies to deliver the new AI solutions.

🔒 ServiceFabric MCP provides an AI-native control surface supporting commercial, open-source, and future models without requiring exclusive facility commitment.

🏥 See All AI validates the technology at the Borton campus, utilizing high-bandwidth connectivity for NVIDIA DGX B200 environments in advanced medical imaging applications.

⚡ The protocol addresses critical enterprise needs for data movement control, policy enforcement, and partner integration that public cloud APIs alone cannot deliver.

📅 ServiceFabric MCP is available immediately to help shorten time-to-deployment for Private AI workloads across the global ecosystem.

Bullish Signals
  • Digital Realty launches a new programmable infrastructure protocol (ServiceFabric MCP) that extends its platform of over 800 data centers, positioning it as a leader in the emerging Private AI market.
  • The company has secured partnerships with major hardware vendors NVIDIA and AMD and software partners ePlus, Lenovo, and Dell to build and validate its AI solutions.
  • A key customer, See All AI, confirms the technology's ability to support high-performance NVIDIA DGX B200 environments for critical medical imaging workloads, validating the platform's technical capabilities.
  • The new protocol addresses a specific market inflection point where enterprises require control over data movement and policy enforcement that public cloud APIs cannot provide.
Full Analysis
Digital Realty Trust (NYSE: DLR) announced the availability of ServiceFabric Model Context Protocol (MCP), an open standard designed to make its global data center infrastructure programmable for Private AI environments. This launch extends Digital Realty's Foundation for AI strategy by introducing patented policy and orchestration technology via its AI Private Exchange (AIPx) architecture, enabling enterprises to securely connect AI workloads across distributed environments. The new protocol is validated across Digital Realty's platform of over 800 data centers and third-party facilities. It supports enterprise deployments across public cloud, network service providers, and bare metal platforms without requiring commitment to a single AI model or exclusive facility usage. The technology facilitates secure Layer 2 and Layer 3 connections with strong authentication, addressing the growing demand for control over data movement and policy enforcement in production AI workloads. Digital Realty has launched these AI solutions with partners including ePlus, Lenovo, and Dell, utilizing infrastructure powered by NVIDIA and AMD technologies. A specific customer validation is provided by See All AI, which utilizes Digital Realty's Borton campus to support its NVIDIA DGX B200 environment for advanced medical imaging systems. The company states that ServiceFabric MCP is available today to help shorten enterprise time-to-deployment for Private AI workloads. Chris Sharp, CTO of Digital Realty, emphasized the strategy to provide foundational infrastructure for sustained AI workloads while enabling flexible scale as demand grows. Mary Johnston Turner of IDC noted that providers combining a global footprint with programmable, agent-ready interconnection are well-positioned to support the next wave of enterprise AI investment. The announcement highlights a shift toward physical infrastructure attributes like power density and sovereign placement being made programmable for the next era of enterprise AI.