Posted On: Sep 15, 2023
Today, we are excited to announce that HAQM EMR on EKS now supports Apache Flink, available in public preview. With this launch, customers who already use EMR can run their Apache Flink application along with other types of applications on the same HAQM EKS cluster, helping improve resource utilization and simplify infrastructure management. For customers who already run big data frameworks on HAQM EKS, they can now let HAQM EMR automate provisioning and management.
Apache Flink is an open-source framework that customers use to run real-time stream processing on high throughput data sources. Prior to today’s launch, customers could run self-managed Apache Flink applications on EKS, but they would be required to manually install and configure Apache Flink which requires in-depth understanding. With today’s launch, customers are only required to install the Flink Kubernetes operator and can use the operator to submit jobs directly to their EKS cluster while letting EMR on EKS manage the lifecycle of the Apache Flink application. This allows customers to share resources across their applications and use a single set of Kubernetes tools to centrally monitor and manage their infrastructure. Additionally, with this launch, customers also get access to key accessibility and monitoring features such as launching Flink application using jars in HAQM S3, monitoring integration with HAQM S3 and HAQM CloudWatch and container log rotation.
To learn more and get started, please visit our Apache Flink section documentation. HAQM EMR on EKS with Apache Flink is now available in preview in regions where HAQM EMR on EKS is currently available.
Oct 5, 2023 - The previous version of this post incorrectly referenced Apache Flink on HAQM EMR on EKS as a fully managed offering. We have now corrected this.