How to modify the block size of my S3 parquet files that are being queried in Athena?

I read that adjusting the block size of the parquet files being queried with Athena can affect and possibly improve the performance of the queries. My parquet files for my database are currently created with DMS (hooked into MS SQL Server as the source). Is it a property of my DMS job I need to change, the parquet files in my S3 bucket themselves, or would I need to create an ETL job with Glue to modify the block sizes after they're created?