Community

NVIDIA Network Operator#

OverviewInstall

Chart Name	Version	App version
network-operator	25.10.0	v25.10.0

The NVIDIA Network Operator manages networking-related components to enable high-performance networking, RDMA, and GPUDirect for workloads running in a Kubernetes cluster. It automates the deployment, configuration, and lifecycle management of NVIDIA Mellanox drivers, device plugins, and supporting services required for advanced networking capabilities.

The Network Operator works in conjunction with the NVIDIA GPU Operator to enable GPUDirect RDMA on compatible systems, ensuring optimal data movement between GPUs and network interfaces while reducing CPU overhead and improving application performance.

Prerequisites#

Deploy k0rdent v1.8.0: QuickStart

Install template to k0rdent#

helm upgrade --install network-operator oci://ghcr.io/k0rdent/catalog/charts/kgst --set "chart=network-operator:25.10.0" -n kcm-system

Verify service template#

kubectl get servicetemplates -A
# NAMESPACE    NAME                            VALID
# kcm-system   network-operator-25-10-0        true

Deploy service template#

apiVersion: k0rdent.mirantis.com/v1beta1
kind: MultiClusterService
metadata:
  name: nvidia-network-operator
spec:
  clusterSelector:
    matchLabels:
      group: demo
  serviceSpec:
    services:
    - template: network-operator-25-10-0
      name: network-operator
      namespace: nvidia-network-operator

Official docs