NVIDIA Infra Controller

C tier on Bare Metal Provisioning SoftwareScore 6.2 · #19 of 29

Android app: Not listed
Free plan: No
Runs on: api, self-hosted

docs.nvidia.com

Summary

NVIDIA Infra Controller (NICo) is an open-source suite of microservices for site-local, zero-trust lifecycle management of bare-metal infrastructure. It automates hardware discovery, firmware checks, DPU provisioning, network isolation, tenant provisioning, and sanitization. During onboarding, it can discover systems through Redfish, validate SKUs, test hardware, and enforce UEFI and BMC firmware baselines. Tenant isolation spans Ethernet, InfiniBand, NVLink, and Spectrum-X. NICo exposes REST and gRPC APIs, an administrator CLI, and a debugging UI; the REST service can run alongside the site controller or centrally in the cloud. Documented integrations include Keycloak-compatible IAM, Prometheus, Grafana, Loki, OpenTelemetry-compatible stacks, and Kubernetes. It is licensed under Apache 2.0. Deployment uses Kubernetes in the managed datacenter, with at least three nodes specified for high availability; dependencies such as PostgreSQL, Vault, Temporal, cert-manager, and External Secrets Operator must be deployed first. NICo does not handle host OS patching, runtime configuration, workload scheduling, physical network underlay design, or UFM deployment and management. Attestation occurs during ingestion and after tenant release, not continuously during tenant use.

Who it is for

NICo is aimed at NVIDIA Cloud Partner infrastructure operators and others managing bare-metal fleets. ISV developers and platform engineers can also integrate with its APIs.

What is good

Automates hardware discovery and firmware validation.
Supports tenant isolation across several network types.
Provides REST and gRPC APIs.
Open source under the Apache 2.0 license.
Includes secure sanitization between tenant uses.

What to know first

Requires a Kubernetes cluster in the managed datacenter.
At least three nodes are specified for high availability.
Host OS patching and runtime configuration remain external.
Attestation is not continuous during tenant use.

Verdict

NICo covers bare-metal onboarding, provisioning, isolation, and sanitization for infrastructure operators. Its Kubernetes dependencies and explicit scope boundaries matter when assessing operational fit.

Compared on bare metal provisioning software

Bare-metal discovery: Yesdocs.nvidia.com
OS image deployment: Yesdocs.nvidia.com
PXE bootstrapping: Yesdocs.nvidia.com
Lifecycle management: Yesdocs.nvidia.com
Hardware inventory: Yesdocs.nvidia.com
API access: Yesdocs.nvidia.com
Out-of-band management: Yesdocs.nvidia.com

Facts

Purpose: NICo is an open-source suite of microservices for site-local, zero-trust bare-metal lifecycle management aimed at NVIDIA Cloud Partners and infrastructure operators.docs.nvidia.com · 3 Oct 2026
License: NICo is open source under the Apache 2.0 license.docs.nvidia.com · 3 Oct 2026
Hardware validation: NICo supports Redfish discovery, SKU validation, hardware testing, and enforcement of UEFI and BMC firmware baselines before hosts become available to tenants.docs.nvidia.com · 3 Oct 2026
Network isolation: It provides tenant isolation across Ethernet, InfiniBand, NVLink, and Spectrum-X using mechanisms including BlueField HBN, UFM P_Key assignment, and NMX-C partition management.docs.nvidia.com · 3 Oct 2026
Security: NICo documents Measured Boot attestation, UEFI lockdown, managed BMC and UEFI credentials, and secure erasure of NVMe storage, GPU memory, and system memory between tenant uses.docs.nvidia.com · 3 Oct 2026
APIs and interfaces: NICo provides REST and gRPC APIs, an administrator CLI, and a debugging UI; its REST API can be deployed alongside the site controller or centrally in the cloud.docs.nvidia.com · 3 Oct 2026
Integrations: Documented integrations include Keycloak and compatible IAM solutions, Prometheus metrics, Grafana, Loki, OpenTelemetry-compatible stacks, and Kubernetes.docs.nvidia.com · 3 Oct 2026
Deployment requirements: The documented architecture deploys NICo microservices on a Kubernetes cluster co-located in the managed datacenter and specifies at least three nodes for high availability.docs.nvidia.com · 3 Oct 2026
External dependencies: NICo depends on services including PostgreSQL, Vault, Temporal, cert-manager, and External Secrets Operator; the documentation says these must be deployed before NICo installation.docs.nvidia.com · 3 Oct 2026
Scope limits: NICo does not manage host OS patching or runtime configuration, workload scheduling, physical network underlay design, or deployment and management of UFM.docs.nvidia.com · 3 Oct 2026
Hardware compatibility: The hardware compatibility list says it is not an exhaustive platform certification or qualification, and directs users to hardware manufacturers for that information.docs.nvidia.com · 3 Oct 2026
Support model: NICo documentation says operating-system management, including patching, configuration, and image generation, is the user's responsibility.docs.nvidia.com · 3 Oct 2026
What it does: NICo is an open-source suite of microservices for site-local, zero-trust bare-metal lifecycle management.docs.nvidia.com · 4 Oct 2026
Hardware onboarding: NICo discovers hardware through Redfish, validates machine SKUs, supports hardware burn-in and testing, and enforces firmware baselines during ingestion.docs.nvidia.com · 4 Oct 2026
DPU management: NICo installs the DPU operating system, provisions HBN, manages DPU firmware, and runs an agent that retrieves desired configuration over gRPC.docs.nvidia.com · 4 Oct 2026
APIs and deployment: NICo exposes REST and gRPC APIs to higher-level systems, and its REST service can run alongside the site controller or centrally in the cloud.docs.nvidia.com · 4 Oct 2026
Attestation limit: Measured Boot attestation occurs during managed-host ingestion and after a host is released by a tenant; NICo says it does not attest hosts continuously during tenant use.docs.nvidia.com · 4 Oct 2026
Hardware limits: NVIDIA says its Hardware Compatibility List identifies hardware currently supported from NICo's perspective and is not an exhaustive platform certification or qualification.docs.nvidia.com · 4 Oct 2026