Posted On: May 15, 2024

Today, AWS announces general availability of CoreDNS autoscaling capabilities for Amazon EKS clusters. This feature allows you to scale capacity of DNS server instances to meet the ever-changing capacity needs of your services without the overhead of managing custom solutions.

Organizations are standardizing on Kubernetes as their compute infrastructure platform to build scalable, containerized applications. Scaling CoreDNS Pods is key to ensure reliable DNS resolution by distributing the query load across multiple instances, and provide high availability for applications and services. With this launch, you no longer need to pre-configure the scaling parameters and deploy a client on each cluster to monitor the capacity and scale accordingly. EKS manages the autoscaling of DNS resources when you use the CoreDNS EKS add-on. This feature works for CoreDNS v1.9 and EKS release version 1.25 and later. For more information about which versions are compatible with CoreDNS autoscaling, visit Amazon EKS documentation. You can benefit from the simplified out-of-box managed option that requires minimal configuration and helps improve the resiliency of your applications. We recommend using this feature in conjunction with other EKS Cluster Autoscaling best practices to improve overall application availability and cluster scalability.

Autoscaling capabilities for CoreDNS Pods are available in all regions where Amazon EKS is available. To get started, visit the Amazon EKS documentation