Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Inter-worker encryption

With Amazon EMR versions 6.15.0 and higher, you can enable mutual-TLS encrypted communication between workers in your Spark job runs. When enabled, EMR Serverless automatically generates and distributes a unique certificate for each worker provisioned under your job runs. When these workers communicate to exchange control messages or transfer shuffle data, they establish a mutual TLS connection and use the configured certificates to verify the identity of each other. If a worker is unable to verify another certificate, the TLS handshake fails, and EMR Serverless aborts the connection between them.

If you're using Lake Formation with EMR Serverless, mutual-TLS encryption is enabled by default.

Enabling mutual-TLS encryption on EMR Serverless

To enable mutual TLS encryption on your spark application, set spark.ssl.internode.enabled to true when creating EMR Serverless application. If you're using the Amazon console to create an EMR Serverless application, choose Use custom settings, then expand Application configuration, and enter your runtimeConfiguration.

aws emr-serverless create-application \
--release-label emr-6.15.0 \
--runtime-configuration '{
  "classification": "spark-defaults", 
  "properties": {"spark.ssl.internode.enabled": "true"}
}' \
--type "SPARK"

If you want to enable mutual TLS encryption for individual spark job runs, set spark.ssl.internode.enabled to true when using spark-submit.

--conf spark.ssl.internode.enabled=true