On-premises devices generate .csv files and can write to an SMB file share. The company wants to analyze the data in AWS using SQL queries; analysts will run queries periodically throughout the day. Which combination of steps will meet these requirements most cost-effectively? (Choose three.)

Deploying an AWS Storage Gateway in S3 File Gateway mode allows on-premises devices to write .csv files to an SMB share, which then transparently stores them in Amazon S3. This provides a cost-effective way to transfer data to AWS. An AWS Glue crawler can then automatically infer the schema of the .csv files in S3 and create a table in the AWS Glue Data Catalog. This catalog entry makes the data queryable. Finally, Amazon Athena is a serverless query service that allows analysts to run SQL queries directly on data stored in S3 using the Glue Data Catalog, paying only for the data scanned. This approach is highly cost-effective for periodic queries compared to maintaining an EMR or Redshift cluster. FSx File Gateway is incorrect because it's designed for FSx for Windows File Server, not S3. EMR and Redshift are less cost-effective for periodic, ad-hoc queries due to their continuous running costs or provisioning overhead.