Fabric and Azure

What Is Microsoft Fabric? Workloads, OneLake and Licensing

Microsoft Fabric in plain terms: its seven workloads, OneLake, how capacity and licensing work, and a worked example from Lakehouse to Power BI report.

Austin Levine

·

Updated

Microsoft Fabric is a single software-as-a-service platform that combines data integration, data engineering, data warehousing, data science, real-time analytics and Power BI reporting, all reading and writing the same underlying storage layer, OneLake. Instead of provisioning and connecting several separate Azure services, a team builds pipelines, models and reports inside one Fabric workspace.

The seven workloads

Fabric is organized into workloads, each a set of item types you create inside a workspace:

Workload

What you build with it

Data Factory

Pipelines and Dataflows Gen2, for moving and transforming data from other sources

Data Engineering

Lakehouses and Spark notebooks, for large-scale processing in code

Data Warehouse

A SQL-based warehouse with a T-SQL query surface

Data Science

Notebooks and models for machine learning against Fabric data

Real-Time Intelligence

Eventstreams, Eventhouses and Activator, for data that arrives continuously rather than on a schedule

Databases

Fully managed operational databases, including SQL database in Fabric

Power BI

Semantic models, reports and dashboards built on data from any of the above

Every workload writes to OneLake, one logical data lake for the whole tenant. A table a data engineer builds in a Lakehouse is available to a Power BI report or a Data Science notebook without exporting or copying it first.

A worked example

A small but complete Fabric flow, using only the free items in a workspace:

  1. Create a Lakehouse in a Fabric workspace.

  2. Build a Data Factory pipeline with a Copy activity that loads a CSV file, or a table from a connected source, into a table in that Lakehouse.

  3. Query the loaded table directly in the Lakehouse's SQL analytics endpoint, a read-only SQL endpoint Fabric creates automatically for every Lakehouse, to confirm the data landed correctly.

  4. Open Power BI Desktop, connect to the Lakehouse, and build a semantic model on top of the table using Direct Lake mode, which reads the Lakehouse's data without a separate Import step.

  5. Publish the report to the same workspace, next to the Lakehouse and the pipeline.

Every step in this flow happens inside one workspace with one set of permissions. Before Fabric, steps 2 and 4 would typically sit in two different products with two different access models. For how Power BI specifically connects to Fabric data, including Direct Lake, see Microsoft Fabric and Power BI.

Capacity and licensing

Fabric is billed through capacity, not per user for the workloads above. Capacity SKUs run from F2 up to F2048, and a trial capacity is available with no Azure subscription required, which is enough to build and test the flow above before committing to a paid capacity. Power BI report viewers still need a Power BI license (Free, Pro or Premium Per User) on top of the workspace's Fabric capacity, except that a Free-license user can view content in a workspace backed by an F64 or larger capacity.

Turning Fabric on for a tenant

A tenant administrator controls whether users can create Fabric items under Tenant settings > Microsoft Fabric > Users can create Fabric items in the Power BI admin portal. This setting can also be delegated to capacity administrators, so it does not have to be a single organization-wide switch. Fabric became generally available in November 2023; most Power BI tenants created since then already have it enabled. For the fuller case for turning it on, see why choose Microsoft Fabric.

FAQs

What is included in Microsoft Fabric?

Seven workloads: Data Factory, Data Engineering, Data Warehouse, Data Science, Real-Time Intelligence, Databases and Power BI, all sharing one storage layer, OneLake.

Do I need to buy anything to try Microsoft Fabric?

No. A Fabric trial capacity is available at no cost and with no Azure subscription, and it supports every workload above except the AI features that require a paid capacity or Copilot seat.

How is Microsoft Fabric licensed?

Through capacity, not per seat, for the data workloads. Capacities run from F2 to F2048. Power BI report viewers separately need a Power BI license unless their workspace runs on an F64 or larger capacity.

Sources

Is your Power BI set up the right way?

1

Book 45 minutes and tell us what you want looked at.

2

We walk through your reports and setup with you, live.

3

Within two working days, you get our plan in writing.

Is your Power BI set up the right way?

1

Book 45 minutes and tell us what you want looked at.

2

We walk through your reports and setup with you, live.

3

Within two working days, you get our plan in writing.

Is your Power BI set up the right way?

1

Book 45 minutes and tell us what you want looked at.

2

We walk through your reports and setup with you, live.

3

Within two working days, you get our plan in writing.

CaseWhen is a Berlin BI consultancy that builds reporting that leaders can trust, on the Microsoft stack: Power BI, Fabric and Azure.

Berlin, Germany

© CaseWhen Consulting GmbH

English

CaseWhen is a Berlin BI consultancy that builds reporting that leaders can trust, on the Microsoft stack: Power BI, Fabric and Azure.

Berlin, Germany

© CaseWhen Consulting GmbH

English

CaseWhen is a Berlin BI consultancy that builds reporting that leaders can trust, on the Microsoft stack: Power BI, Fabric and Azure.

Berlin, Germany

© CaseWhen Consulting GmbH

English