NVIDIA BlueField-4 Powers New Scale-In Network Infrastructure for Agentic AI Factories

Summary

Agentic AI factories connect diverse users, agents, applications, data sources, and storage systems to massively accelerated compute, making dedicated DPU processing essential for line-rate networking, storage, and security. Scale-In evolves north-south access into a coordinated infrastructure domain built with BlueField-4, NVIDIA DOCA, and Spectrum-X Ethernet. BlueField-4 combines programmable control-plane processing with inline data-path engines, a 64-core Grace CPU, LPDDR5X memory, PCIe Gen6 host connectivity, and an 800 Gb/s network interface. The architecture offloads security, tenant isolation, storage protocols, provisioning, and telemetry from host CPUs; the source states that Spectrum-X storage paths can provide up to 1.45x the throughput of off-the-shelf Ethernet, while independent control and observability support more consistent operations as AI infrastructure scales.

Takeaways

1. 1

   BlueField-4 keeps infrastructure processing outside tenant hosts, allowing security policies, service state, and telemetry to be managed without relying on host CPUs or consuming their resources.

2. 2

   BlueField-4 accelerates NVMe-oF, file and object storage protocols over RDMA and TCP, storage virtualization, and data movement; the source reports up to 1.45x more storage throughput than off-the-shelf Ethernet with Spectrum-X.

3. 3

   DOCA Platform Framework provisions, manages, and scales BlueField DPUs as Kubernetes nodes, while DOCA Telemetry exports infrastructure signals for monitoring network traffic, storage access, service health, and utilization.