Forum Discussion
Implementing the right data model for a data product
I don't have experience with the data product concept. However here are my two cents:
I think you need some way to tell your organization that the data product exists. Maybe a data catalog? Purview? Or just use the built in OneLake Data Hub in Fabric.
Anyway I think you need to describe the contents, columns and business context, relationships, etc. of the data in the semantic model. Also relationships to/dependencies on other data products, shared dimensions, etc. Documentation of specific calculation rules which have been applied to the data, if relevant. What are the data sources.
Define who is the data owner and other applicable roles like technical responsible person/team. How to apply for access to the semantic model. Describe whether users are allowed to share the data in the semantic model with others. Refresh frequency. Perhaps some SLA level information.
So I think you need some kind of data catalog to provide all this information.
Some information could potentially be included as metadata inside your semantic model (in various description-attributes of the tables/relationships in the semantic model).
Clear and concise table names and column names in human-understandable language.
Use the labelling features (sensitivity labels) and endorse (certify) the semantic model.
Simplify the data model as much as possible, which makes it easy for the user to understand. Don't have unnecessarily many tables. Denormalize dimension tables. Star schema and multi-fact star schemas is the ideal.
I think you should have some communication channel to tell all users if there are some temporary issues with the data product.
I don't have experience with the data product concept. However here are my two cents:
I think you need some way to tell your organization that the data product exists. Maybe a data catalog? Purview? Or just use the built in OneLake Data Hub in Fabric.
Anyway I think you need to describe the contents, columns and business context, relationships, etc. of the data in the semantic model. Also relationships to/dependencies on other data products, shared dimensions, etc. Documentation of specific calculation rules which have been applied to the data, if relevant. What are the data sources.
Define who is the data owner and other applicable roles like technical responsible person/team. How to apply for access to the semantic model. Describe whether users are allowed to share the data in the semantic model with others. Refresh frequency. Perhaps some SLA level information.
So I think you need some kind of data catalog to provide all this information.
Some information could potentially be included as metadata inside your semantic model (in various description-attributes of the tables/relationships in the semantic model).
Clear and concise table names and column names in human-understandable language.
Use the labelling features (sensitivity labels) and endorse (certify) the semantic model.
Simplify the data model as much as possible, which makes it easy for the user to understand. Don't have unnecessarily many tables. Denormalize dimension tables. Star schema and multi-fact star schemas is the ideal.
I think you should have some communication channel to tell all users if there are some temporary issues with the data product.
Hi pmscorca
We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet.
In case if you have any resolution please do share that same with the community as it can be helpful to others.
Otherwise, will respond back with the more details and we will try to help.Thank you.
