Data is the new app. Rather than building custom applications, modern BI and data science tools allow end users to access and process data and design reports to share findings. To enable this workflow, companies of all sizes are seeking to empower any user, technical or not, to access the data they need. This paradigm has driven the mass adoption of cloud-based data warehouses, lakes, and exchanges to accelerate time to data and simplify data sharing. In turn, a whole new data stack has emerged around cloud compute technologies like Databricks and Snowflake, allowing organisations to break down data silos and easily connect any data tool to any data. However, in the modern era, there is a caveat to data access: regulations such as GDPR, CCPA, and HIPAA require unique data policy management to ensure organisations can control and prove who has access to what data and why. Scaling policy management in the cloud requires a new framework for authoring and enforcing data access controls. Much like the separation of compute from storage in big data, scaling data policy requires the separation of policy from platform.
– The core design patterns of the modern data stack
– Why a new cloud architecture is needed to apply policy at scale
– Lessons from applying such an architecture in real world use cases