You can configure the data object write operation properties that determine how to write data to Snowflake. After you create a Snowflake data object, create a write operation. You can use the Snowflake data object write operation as a target in streaming mappings.

To run streaming mappings with Snowflake as the target, you must specify the private key to authenticate to Snowflake and the Snowflake internal stage in the advanced properties of the Snowflake data object write operation.

When you run streaming mappings to write to Snowflake, the Spark engine writes data to a streaming Snowflake data frame. The data frame in turn writes to the Snowflake internal stage. After the data is available in the Snowflake internal stage, Snowpipe, which is Snowflake's continuous data ingestion service, loads data from the internal stage to Snowflake.

To avoid conflicts while loading data, do not use the stage used by PowerExchange for Snowflake with any other streaming job or to load other files. Do not manually add, modify, or remove files from the internal stage during streaming.