Fabric Persona(s)

With the release of Fabric, the Power BI (Fabric) portal has been updated. In order for a better look and feel, Microsoft has decided to split the areas into Personas. That was so all the artifacts available will not be so overwhelming. Below is the menu that displays from the bottom right of the portal.

Personas

The first 3 are type of applications:
1. Power BI – the previous version of the portal

2. Data Factory – the Fabric default ETL and Orchestration pipeline

3. Data Activator (preview as of Dec 2023) – Fabric alerting tool for code free setup

With any of these selections, you want to make sure you are in the correct Workspace before proceeding. Normally, you are in your My Workspace when logging into the portal.

The next 4 are listed under Synapse.

Data Engineering

This area is for creating a Lakehouse and supporting artifacts like notebooks and pipelines. It is for the Synapse developer or architect moving to Fabric. The Lakehouse is an equivalent to Warehouse we will mention below for the Data Warehouse persona. This is really for those who have been coding in Python (notebooks) and want to use the Delta Table structure supported by OneLake.

Data Science

This area is for creating Machine Learning models and experiments as well as notebooks for coding analytics. This is not a typical area for Power BI users, but does give an option for those hard core analytics users as an entry point to Power BI.

Data Warehouse

This area is for ETL developers that have been using Power BI previously for getting data and transforming it in Dataflows or PowerQuery. It has the no code/less code model for interfacing to Extract, Transform and Load scenarios. The warehouse acts as a Table and File area for querying data with T-SQL rather than coding with a language like Python – think server-less SQL instance.

Real-Time Analytics

Like Data Science, this is an entry point for those wanting real time streaming integrated into Power BI. This is a new area that will have a lot of pain points until fully implemented including a new language KQL.