Getting started with Kubernetes

Before we start digging into how Kubernetes (K8s) works, we should familiarize ourselves with terms that are used in K8s world.

Container — A container or a container image is a lightweight software package that you would want to deploy in your K8s environment. Container image is immutable, standalone executable that can be deployed regardless of the deployment environment. Generally, a container is available in a registry.

Pod — Pod is the smallest deployable unit that is group of one or more containers which use shared network and storage based on a specification which is also a part of Pod.

Node — To run a Pod, a worker machine called “Node” (or Worker Node) is required which can be either virtual machine or a physical one. A node can host several pods and K8s handles scheduling pods automatically. Each node is managed by a Main node which helps in scheduling pods depending on the available resources.

Container Runtime (Docker) — A container runtime enables a pod to run in a node. A docker runtime downloads the image, unpack the image into bundle and run the container from the bundle.

Deployment — Deployment help in creating multiple, identical pods in a node with no unique identities. It runs multiple instances or replicas of the application and replaces any instance automatically in case of application failure or unresponsiveness. K8s deployment controller manages all the deployments.

Kubelet — In order to communicate to the developer or a control plane, kubelet is used. Kubelet lets you interact with a node by executing any execution by interacting with developer or control plane. It also makes sure that containers are running in a pod

Now visualize what we just understood:

In this tutorial, we familiarized ourselves with basic K8s terms. Hopefully, the tutorial helps you to start your journey of K8s.

Keep Learning! Cheers!