Parquet data to AWS Redshift slow

Question

I want to insert data from S3 parquet files to Redshift.

Files in parquet comes from a process that reads JSON files, flatten them out, and store as parq