Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsAmazon EKS Auto Mode is a good fit if you want AWS to manage more of Kubernetes’ data-plane infrastructure and can accept its constraints on node access and lifecycle. It is less suitable if you need SSH or SSM access, heavily customized node operating systems, or full control over scaling components. For Terraform, AWS’s AI/ML walkthrough offers a runnable starting point, but it is a workload-oriented sample—not a universal production configuration.
What EKS Auto Mode changes
With EKS Auto Mode, AWS manages more than the Kubernetes control plane. Its managed capabilities include compute autoscaling, pod and service networking, application load balancing, cluster DNS, block storage, and GPU support. Auto Mode uses Karpenter-based scaling and can consolidate unused capacity. AWS describes the managed capabilities and responsibilities here.
As an Amazon Associate I earn from qualifying purchases.
That does not transfer responsibility for your applications or the whole environment to AWS. You remain responsible for application-container availability, security, and monitoring, as well as your VPC infrastructure and EKS cluster configuration. AWS puts it plainly: “Under EKS Auto Mode, customers continue to maintain responsibility for the application containers, including availability, security, and monitoring.”
Is Auto Mode right for your team?
| Decision area | Auto Mode may fit if… | Consider self-managing if… |
|---|---|---|
| Operations | You want AWS to manage nodes, scaling, and several cluster capabilities. | Your team is prepared to operate more infrastructure components itself. |
| Node access | You can work with immutable, locked-down Bottlerocket variants without direct node login. | You need SSH or SSM access, a customized node operating system, or extensive node-level configuration. |
| Scaling and disruption | The managed scaling, consolidation, node lifecycle, and disruption behavior suit your workloads. | You need to own and tune those mechanisms or require more control over node replacement. |
| IAM and networking | Your IAM roles, trust relationships, VPC, and subnets can meet EKS requirements. | Your existing architecture makes those requirements difficult to satisfy. |
| Cost | Your estimate for the specific region, capacity, storage, load balancing, and observability is acceptable. | You need to compare alternatives under identical workload assumptions before deciding. |
Auto Mode nodes have a maximum lifetime of 21 days and are replaced. Disruption controls can sometimes block upgrades, requiring operator intervention. These lifecycle rules matter for workloads that assume nodes persist or that require tightly controlled maintenance timing. See AWS’s Auto Mode documentation.
Requirements to check before using Terraform
- Kubernetes: Auto Mode requires Kubernetes 1.29 or later and EKS access entries.
- Subnets: Cluster networking requires at least two subnets. The advanced creation guide also expects a qualifying existing VPC and subnets.
- IAM: Confirm the required permissions and trust configuration for the Auto Mode cluster. If you enable built-in node pools, provide a node IAM role.
- Tools: AWS’s Terraform AI/ML walkthrough listed Terraform 1.15.0 or later, kubectl 1.36 or later, AWS CLI 2.27 or later, and jq when its page was checked on October 7, 2026. Check the live walkthrough for updated versions before starting.
- Account and region: Decide which account and region to use, and verify that your VPC and subnets are ready there.
AWS’s advanced cluster creation guide covers IAM, VPC, subnet, and role requirements. The cluster creation documentation gives the minimum Kubernetes version and subnet count.
Set up the AWS Terraform sample
AWS’s AI/ML walkthrough is a concrete starting point if its workload-oriented setup matches your needs. It is not proof that the same configuration is suitable for every production cluster, and its example should not be treated as a production-ready module prescription.
Rank #2
- Review the walkthrough and confirm prerequisites. Check the current tool versions, select the account and region, and prepare the required VPC, subnets, cluster IAM role, and—if using built-in node pools—node IAM role.
- Clone the sample repository. AWS directs readers to clone
aws-samples/sample-eks-docs. - Enter the Auto Mode example directory. From the repository root, run
cd ai-ml/set-up-cluster/terraform/auto-mode. - Initialize Terraform. Run
terraform initin that directory. - Review the planned resources and settings. Verify the selected region, account, network, IAM configuration, node pools, and any workload-specific resources before approving changes.
- Apply the configuration. Run
terraform applyand review the proposed changes before confirming. AWS says deployment takes a few minutes in this tutorial; that is the walkthrough’s description, not a guarantee for every account or environment.
The sample defaults to us-east-2; AWS describes using a region variable to choose another region. Consult the live walkthrough for its current instructions and sample details.
Secure the sample before exposing it
The walkthrough warns that its Grafana load balancer defaults to public HTTP and default credentials. That is a risk in this sample’s Grafana configuration—not a claim about every Auto Mode cluster. Before deploying it, restrict the allowed source CIDR at a minimum and change the Grafana admin password. Consider internal-only access and TLS as well. Review the exposure and credentials before making the endpoint available.
Rank #3
Choosing Auto Mode or a self-managed Karpenter path
AWS’s sample repository has separate terraform/auto-mode/ and terraform/karpenter/ paths. The Auto Mode path relies on AWS-managed components; the self-managed Karpenter path requires you to install and configure more components, including Karpenter and networking add-ons. AWS says switching between those tutorial paths requires destroying the cluster and starting over, so choose a path before deployment rather than treating them as interchangeable.
Auto Mode also supports custom NodePools and NodeClasses alongside its defaults. These can help with instance selection, workload isolation, and ephemeral-storage controls. AWS advises against editing its default NodePools and NodeClasses. The choice between the managed and self-managed paths is about operational ownership, access, and customization; the available guidance does not establish a universal cost winner.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Estimate cost and clean up
The tutorial can create chargeable EKS, GPU, Application Load Balancer, and managed Prometheus resources. There is no single price established for the setup: charges depend on your region, selected capacity, workload, and enabled services. Estimate the resources for your own deployment before applying, and compare options using the same workload assumptions rather than assuming Auto Mode is cheaper.
Recommended Free Tools
When you finish experimenting, destroy tutorial resources you no longer need using the sample’s Terraform workflow. Confirm the account and directory first, and review the proposed destruction before approving it; removing the wrong environment can interrupt workloads or delete resources you intended to keep. AWS’s walkthrough is the reference for the sample’s lifecycle instructions.
What this setup does not establish
AWS’s walkthrough is a useful example, but it does not by itself validate a production design for your organization. The documented setup does not supply an exact regional cost comparison or establish the right Terraform provider and module versions for every use. Pin and verify the provider or module versions you choose, review their current documentation, and validate your configuration for your security, availability, and operational requirements before relying on it in production.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
