A stock trading company uses Amazon Redshift to power the Business Intelligence (BI) specific queries which are run on Redshift. The data engineering team at the company needs to provide the sales team access to a historical trades table whose data is stored in Apache Parquet format in an S3 bucket of the company's data lake. The data engineering team should provide access to only a few specific columns in the historical trades table so that the access does not violate the compliance regulations. Which of the following options should be combined together to build a solution for the given use case? (Select three)