Analytics And Secure Data Sharing With Snowflake

In and of itself, data doesn’t hold any special value. Rather, the insights that are uncovered from data analytics uncovers information and metadata about events, products, and historical facts. While that has added a lot of value to our analysis, occasionally that data could contain confidential aspects such as personally identifiable information (PII data such as social security numbers and mailing addresses), protected health information (PHI data such as health history and medical allergies), and secure fiscal data. As data engineers, it is our ethical and legal responsibility to protect that information.

Database systems have evolved in the way they share information from traditional methods such as EDI and APIs to database views. Unlike ordinary RDBMS base tables, a view is a virtual table that’s usually computed from the database when it is accessed. In general, sharing data via views provides an excellent route for creating dashboards, exporting datasets, integrating with ETL workflows, and supports several business cases. However, with all these methods of sharing data, the chance of data being accessible is high, as the security needs to be set up not only at the database level but also at the access level, allowing for a higher chance of…