What is kubernetes?

Kubernetes (K8s) is an open-source container orchestration platform that help to automate the deployment, scaling, and management of containerized applications.

Why use kubernetes?

• Companies moved from physical servers → VMs → containers → Kubernetes to make applications run faster, more efficient (optimize resources usage), and easier to manage at scale.
• companies are moving from monolithic app to microservices architecture increasing the usage of containers and the need to efficiently manage them in various environments.

Kubernetes architecture

Kubernetes has a Control plane - Worker architecture:

Control plane:

The control plane manages the worker nodes and the pods in the cluster. It contains:

• API Server – Entry point for commands (talks to users & other parts).
• Scheduler – Decides where to run new containers.
• Controller Manager – Handles tasks like scaling & failures.
• etcd – Stores all cluster data (like a database).
• Cloud controller manager - ( for cloud cluster. this is in case kubernetes needs to create ressources in the cloud )

Worker node (nodes):

• Kubelet – Talks to the control plane to run and manage containers.
• Container Runtime – Runs containers (e.g., Docker, containerd).
• Kube Proxy – Manages network communication between containers.

NB: control plane node also has kubelet, kube-proxy and container runtime because it runs some system pods

Pods

Smallest unit in K8s, holds one or more containers.

Kubernetes Plugins or Kubernetes Interfaces

• CRI → Manages containers
• CNI → Handles networking
• CSI → Manages storage.

1. CRI (Container Runtime Interface)

• Enables Kubernetes to use different container runtimes (e.g., containerd, CRI-O, Docker).
• Defines how Kubernetes interacts with the runtime to manage containers.
• Decouples Kubernetes from specific runtimes for flexibility.

2. CNI (Container Network Interface)

• Standard for configuring networking in Kubernetes pods.
• Plugins (e.g., Calico, Flannel, Cilium) handle IP allocation, routing, and network policies.
• Ensures pods can communicate within and outside the cluster.

3. CSI (Container Storage Interface)

• Allows Kubernetes to integrate with external storage systems (e.g., AWS EBS, NFS, Ceph).
• Standardizes how storage is provisioned, attached, and mounted to pods.
• Enables dynamic volume provisioning and snapshots.

Key features of kubernetes

• Autoscaling – Increases or decreases containers based on traffic.
• Self-Healing – Restarts failed containers, replaces unhealthy ones.
• Load Balancing – Distributes traffic across containers for efficiency.
• Rolling Updates & Rollbacks – Updates apps without downtime and rolls back if needed (high availability)
• Service Discovery – Finds and connects services without manual setup (using IP adress or DNS).
• Storage Management – Supports local, cloud, or network storage easily.
• Multi-Cloud & Hybrid Support – Runs across AWS, Google Cloud, Azure, and on-prem.
• Automated Deployments – Uses YAML files for easy and repeatable deployments.
• Secrets & Config Management – Manages configuration and sensitive data securely.
• Resource Optimization – Ensures best use of CPU, memory, and storage (automatic bin packing)
• production Support– Can be used for production environments

K8s cluster setup methods

Note: In this class, we will use kubernetes playgrounds on Killercoda for testing and deploy our production cluster in AWS EKS. The following are just for more information on various methods available.

For learning and local development/testing

• Minikube – Runs a single-node K8s cluster on your laptop. You can check the official documentation for minikube just to get more information.
• Kind (Kubernetes in Docker) – Uses Docker to run lightweight K8s clusters.
• K3s – A lightweight Kubernetes distribution for low-resource systems.
• MicroK8s – A lightweight K8s version from Canonical.
• Killercoda playgroungs for Kubernetes: Killercoda is a platform where you get instant access to a real Linux or Kubernetes environment ready to use. Note: In the cluster setup for this class, you will learn how to use it. We will use it for some practices

For production

1. Manual setup (unmanaged)

• using Kubeadm
• Using kubespray
• Using Kops
• From scratch (complex)

2. Cloud based (managed)

• Amazon EKS – Kubernetes on AWS. Note: This is what we will use in this class
• Google GKE – Kubernetes on Google Cloud.
• Azure AKS – Kubernetes on Microsoft Azure.
• Oracle OKE - Kubernetes on Oracle
• LKE - Kubernetes on Linode
• ...

3. Enterprise Kubernetes (K8s distributions)

• Openshift – Red Hat’s enterprise Kubernetes with extra features.
• Rancher – Multi-cluster Kubernetes management.
• Tanzu - vSphere Tanzu Kubernetes Grid (TKG)
• ...

Interacting with cluster

• Command Line Interface (CLI): using kubectl commands
• User Interface (UI): using the native K8s Dashboard or tools like Lens
• Kubernetes API: Using programmatical access (ex. with python scripts, terraform codes etc.)

How to use this repo

The repo is organized and ordered for you to practice the concepts progressively.

In each folder, you will find a readme file to help you understand the concept and practice.

Note: All the files mentioned in the readme can be directly found in the same folder. Just use the content of those files if no other specification is given.