Which solution will meet these requirements?
Use an Amazon Athena SQL query with the Amazon Athena DynamoDB connector to calculate performance metrics on a recurring schedule.
Use an AWS Glue job with the AWS Glue DynamoDB export connector to calculate performance metrics on a recurring schedule.
Use an Amazon Redshift COPY command to calculate performance metrics on a recurring schedule.
Use an Amazon EMR job with an Apache Hive external table to calculate performance metrics on a recurring schedule.
Explanations:
While Amazon Athena can query data in DynamoDB, it may still place demands on the provisioned throughput during queries, which is not optimal for minimal impact on the table’s capacity.
AWS Glue can export data from DynamoDB to a data lake or other storage options without impacting the provisioned throughput significantly. This allows for batch processing of metrics without affecting live read/write capacity.
Amazon Redshift requires a continuous data load process and does not provide a direct method for querying DynamoDB without impacting the source table, making it unsuitable for this scenario.
Using Amazon EMR can lead to increased load on the DynamoDB table during data processing, which may negatively affect the provisioned throughput, contrary to the requirement for minimal impact.
I classify that the answer is:
Use an AWS Glue job with the AWS Glue DynamoDB export connector to calculate performance metrics on a recurring schedule.