Who is accountable for the data?
Data Mesh rests on four principles: domain ownership, data as a product, a self-service platform and federated governance. Zhamak Dehghani sets them out in her foundational article. It is therefore more than a catalogue or a new database.
Where and how will it be used?
A Lakehouse brings together data lake and analytical warehouse use cases. In Fabric, a Lakehouse supports files and tables. That technical choice does not name the people responsible for their quality or business meaning.
An organization can distribute ownership of data products across domains while using a shared Lakehouse platform. Conversely, installing a Lakehouse does not automatically make the organization a Data Mesh.
Start at the right scale
Our recommendation is to avoid distributing responsibilities faster than teams can take them on. For one initial data product, define an owner, users, an update frequency and a quality commitment. Then observe whether shared responsibility actually improves turnaround and reliability. The organizational model should support useful work rather than become an end in itself.
Sources and features may change. Check the current version when planning your project.
All articles ↗