> For the complete documentation index, see [llms.txt](https://developer.harness.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.harness.io/cloud-cost-management/resources/cacm-architecture-diagrams/kubernetes_autostopping_workflow.md).

# Kubernetes AutoStopping Workflow

<figure><img src="/files/EwMXj8nE0hSWAY7fsf7J" alt=""><figcaption><p>Click to view full size image</p></figcaption></figure>

Just like the diagram shows, Kubernetes AutoStopping is a smart way to manage how programs run. It uses a special tool called an Envoy-based router to control how data moves around. This router is like a traffic director, making sure everything flows smoothly. Let's delve into its workings in more detail.

Here's a breakdown of how it works:

1. **Envoy-Based Router:** This tool sits in the traffic path, watching how data moves through the Kubernetes cluster.
2. **Integration with Existing Ingress:** It's smartly connected to the existing traffic routes, guiding data through the AutoStopping router for analysis.
3. **AutoStopping Controller:** This is like the brain of the operation. It watches the traffic patterns and decides how programs should run based on what it sees.
4. **Deployment of AutoStopping Configuration:** When you turn on AutoStopping, it starts a process where the AutoStopping router, AutoStopping Controller, and other tools are set up inside the Kubernetes cluster.
5. **Essential Tools and Configuration Maps:** Alongside AutoStopping, there are important tools and maps that help keep everything organized and working well, ensuring the AutoStopping system runs smoothly.