A company wants to use a data lake that is hosted on Amazon S3 to provide analytics services for historical data. The data lake consists of 800 tables but is expected to grow to thousands of tables. More than 50 departments use the tables, and each department has hundreds of users. Different departments need access to specific tables and columns. Which solution will meet these requirements with the LEAST operational overhead?