A data engineer extracts monthly training data from Amazon Redshift into Amazon S3. The source schema includes TransactionTimestamp (timestamp), CardName (varchar), and CardNo (varchar). The engineer must (1) remove rows with CardNo = NULL, (2) split TransactionTimestamp into TransactionDate and TransactionTime, and (3) rename CardName to NameOnCard. The solution must minimize infrastructure setup, be automated monthly, and place minimal load on the Redshift cluster. Which solution meets these requirements?

The AWS Glue job is the most suitable solution because it directly addresses all requirements. Glue is a fully managed, serverless ETL service that minimizes infrastructure setup. It can connect to Amazon Redshift as a source, perform the required transformations (filtering nulls, splitting timestamps, renaming fields) using its built-in capabilities, and write the output to S3. Its scheduling feature ensures monthly automation, and as a separate service, it places minimal load on the Redshift cluster. Provisioning an EMR cluster adds infrastructure overhead and management complexity. Running queries from an EC2 instance is a manual process and does not scale well. Redshift Spectrum is designed to query data in S3, not primarily for extracting and transforming data from Redshift to S3 with complex transformations, and scheduling it with Lambda adds unnecessary complexity for this specific task compared to Glue's native scheduling.