A data analytics team runs a petabyte-scale Amazon Redshift cluster for their core business intelligence dashboards. They also have several years of historical raw log data stored as Parquet files in an Amazon S3 data lake. Analysts occasionally need to run ad hoc queries that join the recent data in Redshift with the historical archives in S3, but the team wants to avoid the cost and time of loading all the historical data into the cluster. Which solution meets these requirements with the least operational overhead?