kube-saver: Kubernetes cost estimates and resource right-sizing¶
kube-saver estimates Kubernetes CPU and memory costs from resource requests, metrics-server samples, and configured pricing rates. It runs locally and produces a terminal dashboard, HTML reports, and reviewable resource-change plans.
| Task | Guide |
|---|---|
| Install and generate a first report | Getting started |
| Understand costs, telemetry, and limitations | Cost model and FAQ |
| Find supported commands and flags | CLI reference |
| Select a context or configure pricing | Configuration |
| Grant read permissions | RBAC |
| Review recommendations safely | Safety and trust |
| Translate a local plan into managed manifests | GitOps review workflow |
| Run in a container | Deployment |
| Diagnose connection or metrics problems | Troubleshooting |
| Understand modules and data flow | Architecture |
| Compare workflows | Comparison |
| Inspect portable output formats | Self-contained outputs |
| Review reported environment support | Support matrix |
| Inspect dated test records and limits | Support evidence |
| Prepare a release | Release testing |
| Make your first contribution | Contributing |
Start with installation and a first report, then check the cost model, missing-metrics interpretation, and recommendation limits. Cost reductions in plans are modeled; they do not prove a reduction in cloud billing. Generated HTML can be viewed offline, while scans require Kubernetes API connectivity.
These docs describe the source checkout. Use kube-saver version and
kube-saver --help to identify the behavior of your installed package.