k8s-test-infra

Kubernetes test infrastructure for NVIDIA GPU software — mock GPU environments, CI tooling, and testing utilities.

GPU Mock

Turn any Kubernetes cluster into a multi-GPU environment for testing. No physical NVIDIA hardware required.

# 1. Build the image (no published image yet)
docker build -t gpu-mock:local -f deployments/gpu-mock/Dockerfile .

# 2. Load into KIND
kind create cluster --name gpu-test
kind load docker-image gpu-mock:local --name gpu-test

# 3. Install
helm install gpu-mock deployments/gpu-mock/helm/gpu-mock \
  --set image.repository=gpu-mock \
  --set image.tag=local

After install, deploy a consumer to test:

Consumer	Guide
NVIDIA Device Plugin	Quick Start
NVIDIA DRA Driver	Quick Start

Full documentation: gpu-mock Helm chart README

Mock NVML Library

The underlying CGo-based mock libnvidia-ml.so that powers gpu-mock. Use standalone for local development and CI pipelines.

Document	Description
Overview	What Mock NVML is and how to use it
Quick Start	Build and run in 5 minutes
Configuration	YAML configuration reference
Architecture	System design and components

License

Apache License 2.0 — see LICENSE.