how-it-works.md

How the simulator works

This page describes how the simulator works.

0. starts the simulator.

The simulator server works with the following:

• scheduler
• HTTP server

In advance, the simulator needs to launch etcd, controller-manager and kube-apiserver outside. We recommend using KWOK, see docker-compose.yml to know how we wire things up.

1. users request creating resource.

Users can create resources by communicating with kube-apiserver of KWOK via any clients (e.g. kubectl, k8s client library or Web UI)

2. the scheduler schedules a new pod.

When a new pod is created through kube-apiserver, the scheduler starts scheduling.

3. the results of score/filter plugins are recorded.

Normally, when score/filter plugins are called from scheduler, they will calculate the results and return results to the scheduler. But, in the simulator, custom plugins, that behave as score/filter plugin but records result after calculation, are used in scheduler.

4. the scheduler binds the pod to a node.

The scheduler finally binds the pod to a node if succeeded, or move the pod back to queue if failed.

The result store will notice that the pod has been scheduled/marked as unscheduled by the scheduler and add the scheduling results to the pod's annotation.