Understanding the security requirements of data mesh architecture

In a centralized analytics platform, data access and security are managed centrally. The business decides the policies of who should access data and what data should be private. IT departments implement these policies. This has been the traditional setup in most companies, and there is enough evidence that the business goals are not completely met by IT, resulting in a constant conflict between IT and business. The developers, data engineers, and data scientists get caught in the cross-fire. Enabling a new project is also difficult in such an environment, hindering innovation. Some companies solve this problem by merging a portion of the IT team and the business team under common management, creating concepts such as business IT or shadow IT. Data meshes take a different approach.

A data mesh splits the boundaries of the exchange of data into multiple data products. This provides a unique opportunity to partially...