Using Semantic Dataset / Semantic Data Models in DataFlow / Data Mart

[Anonymous]

My organisation has a lot of PowerBI reports in Premium workspaces maintained by respective business units; each report has a semantic data model /dataset behind it.

I am building a business performance dashboard which presents several KPIs based on data already available in those reports. 

Since the data is already present in those reports/datasets, I want to be able to connect with those datasets. My goal is to extract the data from each data model/dataset and build my own tables (by merging data from individual datasets) to support my whole-of-business performance tracker.

I can connect to the datasets in PowerBI desktop using Composite Model. But it does not allow me to do any further transformations on the tables from the datasets/semantic data models.

I CANNOT connect to datasets using dataflows and/or datamart. I need a way to use the tables in the datasets/data models so that I could take advantage of already cleaned and organised data without having to reconnect to a thousand different sources.

Is there a way around?

[Anonymous]

First issue is to be able to connect to semantic models. Currently I don't see a way of connecting using Data Mart or Data Flow. Is there a way to connect to semantic models using PowerQuery online (in dataflow/datamart)?

Second, I am not trying to change the source data. Whatever transformation I mean to apply will be my local copy (the output of data flow). Are dataflows/datamarts another half baked solution that MS is pushing out?

[Anonymous · Accepted solution]

Hi @Anonymous ,

1. There is no way to connect to semantic models in Dataflow or Datamart

2. Dataflows are self-service ETL (Extract, Transform, Load) tools that allow users to connect to various data sources, transform and clean the data, and then store it in a centralized location. Dataflows can be used to create a single source of truth for an organization's data, which can then be used to create reports and dashboards. Dataflows can be scheduled to refresh automatically, ensuring that the data is always up-to-date.

Datamarts are self-service analytics solutions that enable users to store and explore data that is loaded in a fully managed database. Datamarts provide a simple and optionally no-code experience to ingest data from different data sources, extract transform and load (ETL) the data using Power Query, then load it into an Azure SQL database that's fully managed and requires no tuning or optimization. Once data is loaded into a datamart, you can additionally define relationships and policies for business intelligence and analysis. Datamarts automatically generate a semantic model or semantic model, which can be used to create Power BI reports and dashboards. You can also query a datamart using a T-SQL endpoint or using a visual experience.

For more details, you could read related document: Introduction to dataflows and self-service data prep - Power BI | Microsoft Learn

Best regards,
Community Support Team_Binbin Yu
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[aj1973]

Hi @Anonymous 

No there is no way arround, you can't make changes on tables with live connections to AAS. Changes need to be applied directly in the Datasets.