Kubernetes Architecture
Last updated
Last updated
A virtual or physical machine where the Kubernetes tools are installed. A node is a working machine where the containers will be launched by Kubernetes.
A set of nodes grouped together. If one node within the cluster fails, there are other nodes in the cluster that can go up to ensure the application is still accessible.
A node where the control plane in Kubernetes is installed. The master watches over the nodes in the cluster and is responsible for the orchestration of the nodes.
When installing Kubernetes, here's what you are installing:
API server
front-end for users
interfaces and CLI talk to API server to manage the Kubernetes clusters
etcd
Stores all data used to manage the clusters
e.g. We have a bunch of nodes with a master in the cluster. This ensures there are no conflict between the masters of each cluster.
kubelet
Agent that runs on each node in the cluster
Ensures that the containers are running on the nodes as expected
Container Runtime
The software that runs the containers (in most cases, Docker)
Controller
brain behind the orchestration
responsible for noticing when nodes go down, and making decisions when to bring up new containers
Scheduler
responsible for distributing containers across multiple nodes
searches for newly created containers and assigns them to nodes
Ensures all worker nodes run a copy of a pod
Deployments - consists of one or more pods and replicas of pods
Services - Manages the list of the pods' IP addresses (in order for pods to be able to access other pods)
A ConfigMap is an API object used to store non-confidential data in key-value pairs. Pods can consume ConfigMaps as environment variables, command-line arguments, or as configuration files in a volume.
Resources used to maintain stateful applications
An application that holds the state of the application (if the application goes down, this resource "saves" the data)
e.g. Database
