NVIDIA Infra Controller review
A specialized Kubernetes-based control plane for provisioning and operating GPU and DPU infrastructure.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
NVIDIA Infra Controller (NICo) is an open-source suite of microservices for site-local, zero-trust bare-metal lifecycle management. It is designed for NVIDIA Cloud Partners and infrastructure operators managing AI factory-scale infrastructure, with a particular fit for teams operating GPU and DPU environments. The platform covers hardware discovery, inventory collection, SKU validation, burn-in testing, firmware baseline enforcement, DPU provisioning, tenant OS installation, health monitoring, network isolation, lifecycle transitions, and secure sanitization. It runs as a Kubernetes-based control plane and provides REST and gRPC APIs alongside command-line interfaces.
Its strongest fit comes from the depth of its infrastructure-specific workflow. NICo supports Redfish-based discovery and inventory, out-of-band BMC and UEFI management, PXE/iPXE boot, and BlueField DPU OS, networking, and firmware provisioning. Health monitoring combines Redfish and DPU telemetry, while isolation covers Ethernet, InfiniBand, and NVLink. Tenant release and secure sanitization extend coverage beyond initial deployment into infrastructure reuse and lifecycle operations. Integrations include Kubernetes, PostgreSQL, Vault, Temporal, Prometheus, Grafana, Loki, OpenTelemetry, and Keycloak.
NICo should appeal to infrastructure operators that need an open-source control plane for NVIDIA-focused AI environments and want multiple API options for platform integration. Apache 2.0 licensing supports deployment and extension within those operating models. Its specialization is also the main boundary: teams seeking a broadly applicable bare-metal tool outside NVIDIA GPU and DPU infrastructure may prefer a more general-purpose option. The Kubernetes-based architecture further makes it a fit for organizations already operating Kubernetes as part of their infrastructure platform.
NVIDIA Infra Controller pros and cons
- Where it wins
- Covers discovery, validation, firmware, provisioning, monitoring, and sanitization
- Supports BlueField DPU OS, networking, firmware, PXE/iPXE, and tenant OS images
- Offers REST, gRPC, and CLI access with broad infrastructure integrations
- Where it doesn't
- More specialized than general-purpose bare-metal provisioning platforms
- Requires a Kubernetes-based control plane
- Its scope is centered on NVIDIA AI infrastructure operations
NVIDIA Infra Controller fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update