Agentic AI changes the infrastructure pattern for AI factories. One request can trigger many model calls, tool calls, memory lookups, policy checks, storage accesses, and network transfers before a final answer is produced. As more agents run at once and carry context across steps, users, tools, services, and sessions, infrastructure must move, protect, retrieve, and reuse data fast enough to keep GPUs and CPUs productive.
The NVIDIA BlueField platform brings dedicated, programmable infrastructure processing in the AI factory data path. BlueField offloads infrastructure work from host CPUs, accelerates data movement, enforces policy inline, and enables context reuse. These capabilities help deliver production outcomes such as higher GPU utilization, more predictable latency, stronger isolation, lower cost per token, and more tokens per watt.
It combines purpose-built infrastructure and storage processors with NVIDIA DOCA software for AI factories. NVIDIA BlueField-4 DPUs offload, accelerate, and isolate networking, storage, security, telemetry, and control-plane services from host CPUs while accelerating data movement across GPU compute and CPU compute systems. NVIDIA Vera BlueField-4 STX storage processors power a new class of data platforms for context memory, high-performance storage infrastructure, and secure data services across AI factories.
