en.Wedoany.com Reported - Equinix has launched Equinix Fabric Intelligence, an AI-based interconnect service designed to transform networks from static infrastructure into an autonomous business growth engine, addressing the network challenges posed by generative AI and agentic AI.

The emergence of generative AI and agentic AI is driving disruptive changes in enterprise networks. Traditional networks are static, centralized architectures, while AI workloads are distributed, with data flowing frequently between cloud, data centers, and the edge. Anthony Ho, Director of Product Solutions for Equinix Asia Pacific, noted that when agentic AI begins autonomous communication, the number of connections is expected to swell to 10 times the current level. Agentic AI operates in an "Agent to Agent (A2A)" format, involving repeated autonomous communication between multiple AI models and external tools. The hundreds of milliseconds of latency tolerated by traditional networks cannot meet the demands of this autonomous dialogue, making ultra-low latency below 10 milliseconds and microsecond-level precision essential. Additionally, agentic AI generates unpredictable burst traffic, requiring networks to possess adaptive and dynamic learning capabilities.

Unlike traditional SDN (Software-Defined Networking), which primarily automates connectivity, Equinix Fabric Intelligence is an AI-aware interconnect service designed to achieve network autonomy. This service not only creates connection paths but also enables the network to learn, respond, and optimize performance, security, and cost in real time based on telemetry data. Its core component, the "Fabric Super Agent," allows users to configure networks using natural language in everyday tools such as Slack and Teams, eliminating the need for specialized scripting or manual work orders, thereby reducing deployment time from weeks to minutes.

The solution has already achieved results in multiple customer cases. Continental increased its monthly model training frequency by 14 times through distributed AI models; Zetaris accelerated agentic AI development by 6 times while reducing costs to one-third; AiHPC improved medical dataset movement speed by 22 times via high-bandwidth private connections. Equinix has begun operating a "Distributed AI Hub" based on Fabric Intelligence, aiming to maintain data sovereignty and compliance, connecting and protecting AI traffic, policies, and governance through a single framework. The company is transforming its services into forms that AI agents can directly operate programmatically, such as through the MCP (Model Context Protocol) protocol, driving infrastructure evolution toward a more AI-friendly state.