While moving data from Snowflake to Azure SQL Server or SQL Server the following are some suggested considerations:
1. Using Azure Data Factory, this is slow. Some considerations are the capacity/memory of your Azure SQL instance, Snowflake tier and network bandwidth
2. Writing your own data mover using dot net core - this involves the use of the package Snowflake.Data to read data from Snowflake and ADO to write into Azure SQL / SQL Server. Note that ADO provides a fast way to bulk insert data into SQL Server so this approach is much faster when you have many rows. Be sure to read rows in batches to avoid out of memory exceptions.
3. Using ODBC, JDBC.
4. Using an off the shelf data mover such as Windocks
In all cases, be sure to deal with differences in data types and maximum size limitations that differ between SQL Server and Snowflake. Size limitations apply to maximum varchar and binary sizes. Certain types in SQL Server such as uniqueidentifier are not available in Snowflake.
Download a free data mover from Snowflake to Azure SQL / SQL Server below with a performance of 20 million rows per minute
