Introduction to Amazon EKS

Amazon Elastic Kubernetes Service (EKS) is a managed Kubernetes service that makes it easy to run Kubernetes on AWS without installing, operating, or maintaining your own Kubernetes control plane. If your team already uses Kubernetes or you need the portability and ecosystem that Kubernetes provides, EKS is the AWS service for you.

---

What Is Kubernetes?

Kubernetes (often abbreviated as K8s) is an open-source container orchestration platform originally designed by Google. It automates deploying, scaling, and managing containerised applications.

Kubernetes Architecture

+-----------------------------------------------------------+
|                   Kubernetes Cluster                       |
|                                                           |
|  +---------------------+                                  |
|  |   Control Plane     |                                  |
|  |  +-----------+      |                                  |
|  |  | API Server|      |     +------------------------+   |
|  |  +-----------+      |     |   Worker Node 1        |   |
|  |  +-----------+      |     |  +------+ +------+     |   |
|  |  | Scheduler |      |     |  | Pod  | | Pod  |     |   |
|  |  +-----------+      |     |  +------+ +------+     |   |
|  |  +-----------+      |     |  [kubelet] [kube-proxy] |   |
|  |  | etcd      |      |     +------------------------+   |
|  |  +-----------+      |                                  |
|  |  +-----------+      |     +------------------------+   |
|  |  | Controller|      |     |   Worker Node 2        |   |
|  |  | Manager   |      |     |  +------+ +------+     |   |
|  |  +-----------+      |     |  | Pod  | | Pod  |     |   |
|  +---------------------+     |  +------+ +------+     |   |
|                               |  [kubelet] [kube-proxy] |   |
|                               +------------------------+   |
+-----------------------------------------------------------+

Core Kubernetes Concepts

Concept	Description
Pod	The smallest deployable unit — one or more containers that share networking and storage
Deployment	Manages a set of identical Pods, handling rolling updates and rollbacks
Service	A stable network endpoint that routes traffic to a set of Pods
Namespace	A virtual cluster within a cluster for organising resources
ConfigMap	Stores non-sensitive configuration as key-value pairs
Secret	Stores sensitive data (passwords, tokens) encrypted at rest
Ingress	Manages external HTTP/HTTPS access to services
PersistentVolumeClaim	Requests persistent storage for Pods

---

What Does EKS Manage?

The hardest part of running Kubernetes is operating the control plane — keeping the API server, etcd database, scheduler, and controller manager running, updated, and highly available. EKS manages all of this for you.

Component	Self-Managed Kubernetes	Amazon EKS
API Server	You install, configure, and scale	Managed by AWS across 3 AZs
etcd	You manage backups, availability, and upgrades	Managed by AWS, automatically backed up
Scheduler	You configure and maintain	Managed by AWS
Controller Manager	You configure and maintain	Managed by AWS
Worker Nodes	You provision and manage	You manage (EC2) or use Fargate
Kubernetes upgrades	You perform the upgrade process	AWS handles control plane upgrades; you upgrade nodes
High availability	You configure multi-AZ	Built in — control plane spans 3 AZs

---

EKS Node Types

EKS offers three options for running worker nodes:

1. Managed Node Groups

AWS provisions and manages EC2 instances for you:

aws eks create-nodegroup \
  --cluster-name my-cluster \
  --nodegroup-name my-nodes \
  --node-role arn:aws:iam::123456789012:role/eksNodeRole \
  --subnets subnet-aaa subnet-bbb \
  --instance-types t3.medium \
  --scaling-config minSize=2,maxSize=10,desiredSize=3

Advantages:

• Automatic provisioning and lifecycle management
• Rolling updates for AMI and Kubernetes version upgrades
• Integration with Auto Scaling groups

2. Self-Managed Nodes

You create and manage your own EC2 instances and register them with the EKS cluster.

When to use:

• Custom AMIs with specific software
• GPU instances with special drivers
• Full control over the node lifecycle

3. Fargate Profiles

Run Pods on Fargate without managing any nodes at all: