Cisco Unified Edge platform targets real-time AI at the edge
Cette publication existe aussi en Français
Cisco has unveiled Cisco Unified Edge, a new integrated edge computing platform designed to run real-time AI inferencing and agentic workloads directly where data is generated. The company introduced the system during its Partner Summit in San Diego, positioning it as a key building block for enterprise-scale AI deployments.
The announcement highlights how edge compute architectures are becoming essential for managing rapidly growing AI workloads, especially in distributed industrial, retail, healthcare, and telecom environments.
Bringing data center-scale AI to the edge
As AI use shifts from centralized model training to real-time inference, organizations are finding that traditional data center architectures struggle to keep up. According to Cisco, AI agents can generate network traffic as much as 25× higher than typical chatbot workloads, pushing the need for compute and decision-making closer to where data originates.
“Today’s infrastructure can’t meet the demands of powering AI at scale,” said Jeetu Patel, President and Chief Product Officer at Cisco. “As AI agents and experiences proliferate, they will naturally emerge closer to where customers interact and decisions are made – the branch office, retail store, factory floor, stadium, and more. That’s where compute needs to live. With our Unified Edge we’re making it easier to power AI in the real world with flexible, secure systems that are simple to deploy, operate, and scale as demand grows.”
Unified Edge combines compute, networking, storage, and security in a single modular platform. It supports both CPU and GPU configurations and integrates with Cisco’s SD-WAN, Intersight cloud-based management, and observability platforms such as Splunk and ThousandEyes. The system supports zero-touch deployment and fleetwide operations without requiring on-site specialists.
Security and modularity for distributed AI
Cisco builds security throughout the system, using zero-trust architecture, tamper-proof hardware features, and consistent policy enforcement from device to application. This is particularly relevant as the attack surface expands when AI workloads move closer to the physical environment.
A broad partner ecosystem—including Intel, Verizon, Rockwell Automation, and World Wide Technology — helped shape the platform. Intel highlighted the role of its Xeon 6 SoC as a foundation for high-throughput, low-latency edge applications, while Verizon and Rockwell pointed to the growing competitive need for real-time decision-making in manufacturing and service operations.
Cisco Unified Edge platforms are available to order now, and Cisco plans to make them generally available by the end of the year.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
