DRAForge can be deployed in two primary profiles: a Demo Profile for rapid local exploration, and a Production Profile for robust deployment on a Kubernetes cluster.
The Demo profile is optimized for local exploration using a kind cluster with the DRA feature gate enabled. This profile is not suitable for production use as it uses local builds and configurations.
git clone https://github.com/oaslananka/draforge.git
cd draforge
bin/ directory.
task build
kubectl apply -f deploy/crds/simulateddevicepool-crd.yaml
kubectl apply -f examples/scenarios/basic-gpu.yaml
./bin/draforge tui
The Production profile is designed for deploying DRAForge into a remote Kubernetes cluster (e.g., DigitalOcean Kubernetes). This method uses Helm to manage the application lifecycle.
kubectl configured with cluster administrator access.# From the cloned repository root:
cd deploy/helm/draforge
draforge-system).
helm install draforge deploy/helm/draforge \
--namespace draforge-system \
--create-namespace
kubectl get pods -n draforge-system
kubectl get svc -n draforge-system
kubectl port-forward svc/draforge-api-server -n draforge-system 8080:80
Note: Ensure proper ingress and authentication are configured for long-term access, as the dashboard is unprotected by default.