Fabric Implementation Guide · Part 8 of 10

Synapse & Power BI Premium to Fabric Migration

Execute seamless migrations from legacy Power BI Premium (P1-P5) and Azure Synapse Dedicated SQL Pools to Fabric F-SKUs and Unified Warehouses.

Target Audience: Migration Leads, Enterprise Architects, and Database Administrators
Estimated Read Time: 13 min

01Power BI Premium (P SKU) to Fabric (F SKU) Cutover

Microsoft retired legacy Power BI Premium P-SKUs in favor of Fabric F-SKUs (P1 maps directly to F64). Transitioning to F64 maintains free viewer report distribution while granting full access to Fabric Data Factory, Lakehouses, Warehouses, and Real-Time Analytics engines.

Implementation & Verification Checklist

  • Inventory active Power BI Premium P1-P5 capacities and contract expiration dates
  • Verify that target capacity is F64 or higher to preserve free viewer licensing eligibility
  • Migrate import semantic models to zero-copy OneLake DirectLake mode
  • Audit PPU (Premium Per User) licenses and downgrade eligible accounts to standard Pro ($10/mo)

02Synapse Dedicated SQL Pool to Fabric Warehouse

Migrate Azure Synapse Dedicated SQL Pools and ADF pipelines to Fabric T-SQL Warehouses and Data Factory. Convert T-SQL schemas, stored procedures, and views while replacing proprietary table distribution keys with Fabric's automated V-Order Delta Parquet engine.

Implementation & Verification Checklist

  • Inventory Synapse Dedicated SQL Pools, pipeline orchestration, and linked services
  • Convert T-SQL DDL schemas and migrate data to Fabric Warehouse Delta tables
  • Run dual-system parallel reconciliation to verify row counts, aggregates, and query SLAs
  • Decommission legacy Synapse resources upon formal business sign-off

03Azure Data Factory (ADF) & Mapping Data Flows Migration

Ready to move your ADF workloads to Fabric? The migration experience is designed to be incremental, guided, and low risk. From the ADF UX, launch the built-in migration flow to assess pipeline activity readiness, copy linked service metadata directly to Fabric connections, and upgrade pipelines to Fabric (preview) without rewriting core logic. Existing Mapping Data Flows should be reviewed for Spark compatibility.

Implementation & Verification Checklist

  • Launch the built-in ADF assessment tool to flag pipelines needing manual review
  • Map legacy linked services to dynamic Fabric cloud and gateway connections
  • Migrate pipelines selectively in phased waves while continuing to run ADF during the validation phase
  • Verify connector parity and manually re-author complex Mapping Data Flows in Fabric Spark engines