# OpenCost — Real-Time Kubernetes Cost Monitoring

> Monitor and allocate Kubernetes infrastructure costs in real time. A CNCF sandbox project that gives teams visibility into cluster spend by namespace, pod, and label.

## Install

Save as a script file and run:

# OpenCost — Real-Time Kubernetes Cost Monitoring

## Quick Use
```bash
helm repo add opencost https://opencost.github.io/opencost-helm-chart
helm install opencost opencost/opencost --namespace opencost --create-namespace
kubectl port-forward -n opencost svc/opencost 9090:9090
# Open http://localhost:9090 for the cost dashboard
```

## Introduction
OpenCost is a CNCF sandbox project that provides real-time cost monitoring for Kubernetes clusters. It allocates infrastructure spend to namespaces, deployments, pods, and labels so engineering and finance teams can see exactly where cloud dollars go.

## What OpenCost Does
- Tracks CPU, memory, storage, and network costs per Kubernetes workload
- Allocates shared costs like control plane and load balancers across tenants
- Integrates with cloud billing APIs from AWS, GCP, and Azure for accurate pricing
- Exposes a REST API and Prometheus metrics for cost data consumption
- Provides a built-in web dashboard for quick cost exploration

## Architecture Overview
OpenCost runs as a single Go pod that queries the Kubernetes API for resource usage and correlates it with cloud provider pricing data. It pulls on-demand and spot pricing from AWS Cost and Usage Reports, GCP BigQuery billing exports, or Azure rate cards. The cost model is exposed as Prometheus metrics and a JSON API, making it easy to build Grafana dashboards or feed data into FinOps platforms.

## Self-Hosting & Configuration
- Deploy via Helm chart into any Kubernetes cluster running Prometheus
- Configure cloud provider integrations by setting billing API credentials
- Set custom CPU and memory pricing for on-prem or bare-metal clusters
- Adjust shared cost allocation weights via the cost model configuration
- Expose the API externally or scrape metrics with an existing Prometheus stack

## Key Features
- CNCF-backed open standard for Kubernetes cost monitoring
- Real-time cost allocation by namespace, label, controller, and pod
- Native Prometheus metrics integration for alerting and dashboards
- Multi-cloud pricing support including spot and reserved instances
- Lightweight single-pod deployment with no external database required

## Comparison with Similar Tools
- **Kubecost** — Commercial product built on top of OpenCost; OpenCost is the free open-source engine
- **Infracost** — Estimates IaC costs before deploy; OpenCost monitors live runtime spend
- **Cloud provider cost dashboards** — Aggregate billing only; OpenCost maps costs to Kubernetes objects
- **Prometheus + custom queries** — Requires building cost models manually; OpenCost provides them out of the box

## FAQ
**Q: Is OpenCost the same as Kubecost?**
A: OpenCost is the open-source cost allocation engine that Kubecost originally created and donated to the CNCF. Kubecost adds a commercial UI and advanced features on top.

**Q: Does it work with on-prem Kubernetes clusters?**
A: Yes. You can configure custom CPU, memory, and storage rates for clusters without cloud billing APIs.

**Q: What Prometheus setup is needed?**
A: OpenCost works with any Prometheus-compatible stack. It can run its own bundled Prometheus or connect to an existing one.

**Q: Can I see costs per team or product?**
A: Yes. Use Kubernetes labels and namespaces to group costs, then query by those dimensions via the API or dashboard.

## Sources
- https://github.com/opencost/opencost
- https://www.opencost.io/docs

---
Source: https://tokrepo.com/en/workflows/8d0547be-3974-11f1-9bc6-00163e2b0d79
Author: Script Depot