Amazon EMR Serverless adds larger worker sizes
Amazon EMR Serverless now supports larger worker configurations, offering up to 32 vCPUs and 244 GB of memory. This enhancement allows for running more compute and memory-intensive workloads, potentially improving runtime performance and reducing costs for Spark and Hive jobs. The feature is available in all AWS Regions where EMR Serverless is supported, and is recommended for shuffle-heavy, data-skewed, or in-memory caching workloads.
- →Support for larger worker configurations
- →Availability and documentation
Features (1) ›
- Support for larger worker configurations
Amazon EMR Serverless now offers larger worker configurations with up to 32 vCPUs and 244 GB of memory, an increase from the previous maximum of 16 vCPUs and 120 GB of memory. These larger workers are recommended for compute and memory-intensive Spark and Hive workloads to improve runtime performance, reduce data transfer inefficiencies, mitigate out-of-memory failures, and boost job performance by allowing more data to be cached in memory.
Notes (1) ›
- Availability and documentation
Larger worker configurations for Amazon EMR Serverless are now available in all AWS Regions where the service is offered. Further details on worker configurations can be found in the EMR Serverless documentation.
https://aws.amazon.com/about-aws/whats-new/2026/07/amazon-emr-serverless/