Skip to main content

Kubernetes Architecture

Kubernetes follows a control plane–worker node architecture, where the control plane manages the cluster and worker nodes run your workloads.

Kubernetes Cluster Architecture Figure: Simplified Kubernetes cluster architecture.

Control Plane Components​

The control plane makes global decisions about the cluster and ensures the system's desired state.

  • API Server — Acts as the front-end for Kubernetes, exposing the Kubernetes API to users and components.
  • etcd — A highly available key-value store that holds all cluster data.
  • Scheduler — Assigns Pods to nodes based on resource availability and constraints.
  • Controller Manager — Runs background processes (controllers) that maintain cluster health and reconcile actual vs. desired states.

Worker Node Components​

Each node runs the components needed to host and manage Pods.

  • kubelet — Communicates with the control plane and ensures containers are running as expected.
  • Container Runtime — Executes containers (e.g., containerd, CRI-O).
  • kube-proxy — Maintains network rules and manages communication between Pods and services.

Amazon EKS Architecture​

Amazon Elastic Kubernetes Service (EKS) is a managed Kubernetes service that simplifies cluster operations.
It takes care of control plane management, upgrades, and high availability, so you can focus on your workloads.

With EKS, you can:

  • Deploy applications faster with less operational overhead
  • Scale seamlessly to handle changing workloads
  • Enhance security using AWS IAM and managed updates
  • Choose your compute model — traditional EC2 nodes or serverless with EKS Auto Mode

Shared Responsibility Model​

In Amazon EKS:

  • AWS manages the control plane — including the API Server, etcd, scheduler, and controllers.
  • You manage the worker nodes — EC2, Fargate, or hybrid options where your applications run.
  • AWS services integrate natively — including load balancers, IAM roles, VPC networking, and storage.

Amazon EKS Architecture Figure: Amazon EKS architecture and integration with AWS services.

Key Design Principles​

Understanding these principles will help you work more effectively with Kubernetes:

Control Plane vs. Worker Nodes​

  • Control plane components (API Server, etcd, Scheduler, Controller Manager) handle cluster-wide decisions and state management
  • Worker nodes (kubelet, container runtime, kube-proxy) focus on running and networking your applications
  • This separation allows for scalable, resilient cluster operations

EKS Advantages​

  • Reduced operational burden — AWS manages control plane complexity, patching, and high availability
  • Native AWS integration — Seamless connectivity with VPC, IAM, Load Balancers, and other AWS services
  • Flexible compute options — Choose between EC2, Fargate, or Auto Mode based on your workload needs

Core Concepts​

  • Declarative configuration — Define desired state; Kubernetes controllers work to achieve it
  • API-driven — All interactions go through the Kubernetes API for consistency and auditability
  • Extensible — Custom resources and controllers allow you to extend Kubernetes functionality