Using Delta Lake with Amazon EMR on EKS

Delta Lake is an open-source storage framework for building a Lakehouse architecture. The following shows how to set it up for use.

To use Delta Lake with Amazon EMR on EKS applications

1. When you start a job run to submit a Spark job in the application configuration, include the Delta Lake JAR files:

   --job-driver '{"sparkSubmitJobDriver" : {
         "sparkSubmitParameters" : "--jars local:///usr/share/aws/delta/lib/delta-core.jar,local:///usr/share/aws/delta/lib/delta-storage.jar,local:///usr/share/aws/delta/lib/delta-storage-s3-dynamodb.jar"}}'

   Note

   Amazon EMR releases 7.0.0 and higher uses Delta Lake 3.0, which renames delta-core.jar to delta-spark.jar. If you use Amazon EMR releases 7.0.0 or higher, be sure to use the correct file name, such as in the following example:

   --jars local:///usr/share/aws/delta/lib/delta-spark.jar

2. Include Delta Lake additional configuration and use AWS Glue Data Catalog as your metastore.

   --configuration-overrides '{
           "applicationConfiguration": [
           {
             "classification" : "spark-defaults", 
             "properties" : {
               "spark.sql.extensions" : "io.delta.sql.DeltaSparkSessionExtension", 
               "spark.sql.catalog.spark_catalog":"org.apache.spark.sql.delta.catalog.DeltaCatalog",
   "spark.hadoop.hive.metastore.client.factory.class":"com.amazonaws.glue.catalog.metastore.AWSGlueDataCatalogHiveClientFactory" 
              }
           }]}'