A data scientist needs to explore roughly 500 GB of historical inventory data stored as CSV files in an Amazon S3 data lake. They want to use SQL for exploration and the company wants to minimize cost and operational overhead. Which approach satisfies these requirements with the least operational management?

The correct approach is to use AWS Glue to crawl the S3 bucket and populate the AWS Glue Data Catalog, then query the data with Amazon Athena. This is the most cost-effective and operationally simple solution for ad-hoc SQL queries on data in S3. Athena is serverless, meaning there's no infrastructure to provision or manage, and you pay only for the data scanned. AWS Glue Data Catalog provides a managed metadata repository that Athena can directly leverage. Provisioning an Amazon EMR cluster involves managing EC2 instances, which increases operational overhead and cost. While Redshift can query S3 data, provisioning a Redshift cluster (either for ingestion or external tables) introduces significant operational overhead and cost for a 500 GB ad-hoc exploration task, as Redshift is designed for analytical workloads requiring dedicated compute.