SSIS Best Practices with Amazon Redshift

Customers coming from SQL Server might first start with an Amazon Redshift pilot:

1. Create their first Redshift cluster (and secondary AWS things like AWS VPC subnet groups, Amazon S3 buckets, etc.).
2. Load some sample data and get familiar with Redshift fundamentals (COPY command, zone mapping, etc.)
3. Hook up their favorite BI tool to check for compatibility.

As they progress, they can load a bulk copy of data from their existing MSSQL database using AWS DMS so they can experiment with their existing data model, see how it performs in Redshift, find out changes in data types, changes to their queries (especially things like PIVOTs), etc.

Finally, when they are ready to start integrating their data sources directly into Redshift, they might find they don't need SSIS and they can explore cloud-native services such as Glue, or Marketplace offerings like Matillion. If there is a hard requirement for SSIS, I would recommend a 3rd party plugin called ZappySys. It handles the Redshift loading best practices (i.e. staging data in S3, using COPY commands on multiple files in S3, etc.).

Hope this helps!