Forum Discussion
How using or consuming a data product/data mesh in Fabric
- Anonymous2 years ago
Hi pmscorca,
After finish data integrate, transform and data cleanup, you can store these data in Lakehouse or data warehouse.
You can use above source to analytic and output the result. (Use/create semantic model based on tables/views from the SQL analytics endpoint or data warehouse and use this model to create reports with calculated field and visualizations to consume them)
In addition, you can also use data science to input and analyze these records and get the insights of the defined questions.
Regards,
Xiaoxin Sheng
Hi pmscorca that will depend on what Fabric items are your "data products". Are they semantic models? Warehouse? Lakehouse etc?
Then it's a case of the connectivity methods that the Fabric item allows will dictate how you provide access. Within the Fabric ecosystem you would implement security across those items to ensure only the relevant users have access. If you're using endpoints (SQL/XMLA) then you'd still need to set up permissions.
Purview has the concept of creating data products which you can assign Fabric items too, if you want/intend to use Purview.
- pmscorca2 years ago
Post Prodigy
Hi, thanks for your reply.
For now, my data product is saved on a warehouse, but it could be saved on a lakehouse.
A data product could not necessarily be consumed by a BI report and so it isn't needed to create a semantic model; in this sense, Fabric represents a data platform and not an end-to-end service to deliver a BI solution.
F.e. a data product could serve for another data product. It occurs to think to the data mesh approach.
Moreover, are the Purview data product and the data product following the data mesh approach the same concepts?
Thanks
- Anonymous2 years agoNot applicable
HI pmscorca,
In fact, it works as the official document descriptions:
Microsoft Fabric is an end-to-end analytics and data platform designed for enterprises that require a unified solution.
It centralizes data storage with One Lake, operating on a Software as a Service (SaaS) model, and integrates separate components into a cohesive stack.Regards,
Xiaoxin Sheng