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).

DynamoDB zero-ETL integration with Amazon SageMaker Lakehouse

Setting up an integration between the DynamoDB table and Amazon SageMaker Lakehouse require prerequisites such as configuring IAM roles which Amazon Glue uses to access data from the source and write to the target, and the use of KMS keys to encrypt the data in intermediate or the target location.

Prerequisites before creating a DynamoDB zero-ETL integration with Amazon SageMaker Lakehouse

To configure a zero-ETL integration with an DynamoDB source, you need to set up a Resource-Based Access (RBAC) policy that allows Amazon Glue to access and export data from the DynamoDB table. The policy should include specific permissions like ExportTableToPointInTime, DescribeTable, and DescribeExport with conditions restricting access to a specific Amazon Web Services account and region. See, Configuring an Amazon DynamoDB source for more information.

Point-in-time recovery (PITR) must be enabled for the table, and you can apply the policy using Amazon CLI commands. The policy can be further refined by specifying the full integration ARN for more restrictive access control. For more information, see Prerequisites for setting up a zero-ETL integration.

Viewing CloudWatch metrics for integration

Once an integration completes, you can see these CloudWatch metrics and EventBridge notifications generated in your account for each Amazon Glue job. For more information, see Monitoring an integration.