Getting Started
Prerequisites
- A Kubernetes cluster (1.27+) with managed node pools (EKS, GKE Standard, AKS)
- Helm 3.x
INFO
Gerty v1 supports clusters with managed node pools. GKE Autopilot and AWS Fargate support is on the roadmap.
Install
bash
helm repo add gerty https://gerty-labs.github.io/gerty
helm install gerty gerty/gertyVerify
Check that the agent and server pods are running:
bash
kubectl get pods -l app.kubernetes.io/name=gertyYou should see a gerty-agent pod on each node (DaemonSet) and one gerty-server pod.
First Report
Generate a cluster-wide efficiency report:
bash
gerty reportThis shows waste metrics across all namespaces, with workload classification (steady, burstable, batch, idle) and potential savings.
Next Steps
- Configuration - customise agent intervals, SLM, integrations
- CLI Reference - all commands and flags
- Architecture - how the components fit together