Posted On: Jul 24, 2023

We are excited to announce that Amazon EMR has launched two new capabilities that enhances the scaling experience for Amazon EMR on EC2 clusters: a new retry mechanism for faster scaling of your Amazon EMR on EC2 clusters running Presto or Trino; and faster scale-down of Amazon EMR on EC2 clusters by enforcing the data redundancy requirements. These capabilities are automatically enabled for clusters running Amazon EMR 6.12 or higher releases and no action is needed from your end.

Amazon EMR offers you the capability to scale your cluster either manually or by using Automatic Scaling. You can also use the Amazon EMR managed scaling feature to automatically resize your cluster based on workload and utilization. With today’s launch, Amazon EMR adds a new retry mechanism to the scaling workflow for Amazon EMR on EC2 clusters running Presto or Trino. Specifically, this enhancement improves cluster utilization by enabling your cluster to scale up and down faster. Additionally, we also enhanced the scale-down for Amazon EMR on EC2 clusters by ensuring that we don’t scale-down the core nodes below the HDFS replication factor configured for the cluster. This helps improve cluster utilization with faster scale-down of your Amazon EMR on EC2 clusters.

These enhancements are now generally available in all commercial AWS Regions, including the AWS GovCloud (US) Regions, where Amazon EMR on EC2 is available. For more information, please visit the Amazon EMR 6.12.0 release notes.