Course contentsShow
← Cloud and Infrastructure
Lesson 1465 of 2,98930. Managed Kubernetes ServicesFree lesson

Creating Managed Node Groups

Configuring managed node groups with instance types, scaling settings, and launch templates.

Creating Managed Node Groups

You'll learn: How to configure managed node groups in Amazon EKS with the right instance types, scaling parameters, and launch templates to run your containerized workloads efficiently.

What Are Managed Node Groups?

A managed node group is Amazon EKS's automated way of provisioning and lifecycle-managing EC2 instances that serve as worker nodes in your cluster. Think of it like hiring a property management company instead of being a landlord yourself—AWS handles the provisioning, updates, and termination of nodes while you focus on configuration.

Core Configuration Elements

When creating a managed node group, you define three key areas:

Instance Types

Specify which EC2 instance types to use (e.g., t3.medium, m5.large). You can configure a single type or multiple types for flexibility. This determines the CPU, memory, and network capacity available for your pods.

Scaling Settings

Define the minimum, maximum, and desired number of nodes. For example:

  • Minimum: 2 nodes (ensures availability)
  • Maximum: 10 nodes (cost control)
  • Desired: 3 nodes (initial capacity)

These parameters work with Cluster Autoscaler to grow or shrink your node pool based on demand.

Launch Templates (Optional)

Launch templates let you customize node behavior beyond basic settings—like attaching additional security groups, specifying custom AMIs, adding user data scripts, or configuring block device mappings. This gives you fine-grained control while maintaining managed lifecycle benefits.

The Creation Process

You create managed node groups through the AWS Console, AWS CLI, or infrastructure-as-code tools. EKS automatically handles joining nodes to your cluster, applying necessary IAM roles, and configuring the AWS CNI plugin.

Key Takeaway: Managed node groups combine AWS automation with your configuration choices for instance types, scaling bounds, and optional launch templates—giving you production-ready worker nodes without manual setup complexity.