Transform + Fabric

TL;DR

Transform workloads can produce richer data. Azure SQL DB mirroring to Fabric combines cloud-native data models with governed analytics, ML, and real-time intelligence when the source schema, permissions, and network design are ready.

Transform applications can produce richer, more granular data than their legacy predecessors. With Azure SQL Database mirroring to Fabric, that data flows directly into the unified data platform — ready for analytics, AI, and cross-system insights.

Azure SQL DB in Fabric

Like SQL MI Mirroring, Azure SQL Database can mirror its data into Fabric’s OneLake. But because Transform applications are typically more modern in their data patterns, the Fabric integration unlocks additional capabilities.

%%{init: {'theme':'neutral'}}%%
graph LR
  classDef azure   fill:#0078d4,stroke:#005a9e,color:#fff
  classDef onelake fill:#742774,stroke:#5a1e5a,color:#fff
  classDef bi      fill:#fde8f9,stroke:#742774,color:#3a003a
  SQLDB[("Azure SQL Database<br/>Cloud-Native")]:::azure
  ONELAKE(["OneLake<br/>Fabric Data Lake"]):::onelake
  PBI["Power BI<br/>Reports & Dashboards"]:::bi
  ENG["Data Engineering<br/>Spark / Notebooks"]:::bi
  SCI["Data Science<br/>ML Models"]:::bi
  RTI["Real-Time Intelligence"]:::bi
  SQLDB -->|"Mirroring (near real-time)"| ONELAKE
  ONELAKE --> PBI
  ONELAKE --> ENG
  ONELAKE --> SCI
  ONELAKE --> RTI

Why Transform + Fabric is More Powerful

Modernized applications produce better data. Combined with Fabric, this creates a compounding advantage:

Stabilize + Fabric	Transform + Fabric
Mirrors existing database schemas as-is	Modern schemas optimized for analytics
Batch-oriented application data patterns	Event-driven, real-time data streams
Analytics on legacy data structures	Analytics on cloud-native data models
BI dashboards and reports	BI + ML + real-time intelligence

The Unified Data Estate

Whether a customer follows Stabilize, Transform, or both, Fabric can become a governed destination for supported operational and analytical data:

%%{init: {'theme':'neutral'}}%%
graph TB
  classDef azure   fill:#0078d4,stroke:#005a9e,color:#fff
  classDef onelake fill:#742774,stroke:#5a1e5a,color:#fff
  classDef bi      fill:#fde8f9,stroke:#742774,color:#3a003a
  subgraph h1["Stabilize"]
    SQLMI[("SQL Managed Instance")]:::azure
  end
  subgraph h2["Transform"]
    SQLDB[("Azure SQL Database")]:::azure
  end
  OL(["OneLake"]):::onelake
  ANALYTICS["Unified Analytics<br/>BI · ML · Real-Time"]:::bi
  SQLMI -->|"Mirroring"| OL
  SQLDB -->|"Mirroring"| OL
  OL --> ANALYTICS
  style h1 fill:#e6f3ff,stroke:#0078d4
  style h2 fill:#e6f3ff,stroke:#0078d4

Fabric is the destination, not the detour

This is why Fabric is central to the modernization story when analytics, AI, and cross-system insight are part of the strategy. The customer can standardize governance and consumption patterns instead of creating a separate analytics operating model for each workload.

The Strategic Payoff

For the customer, this means:

• One governed analytics platform — Fewer duplicated extracts, pipelines, and analytics silos per application
• Trusted data products — Mirrored and shortcutted data becomes reusable, governed data products consumed across the organization for analytics and AI
• AI-ready by design — Data in OneLake can support machine learning, Copilot integrations, and advanced analytics once ownership, quality, and access controls are in place
• Governed by design — Microsoft Purview provides unified data governance, classification, and policy enforcement across the entire data estate — OneLake, Azure, on-premises, and third-party sources
• Incremental value — Stabilize workloads contribute data to Fabric today; as they evolve to Transform, the data gets richer — but the platform is already in place

Mirroring is not security propagation

Azure SQL Database row-level security, object or column permissions, and dynamic data masking are not propagated into Fabric OneLake. Microsoft Purview Information Protection sensitivity labels from Azure SQL Database are not cascaded into Fabric either. Recreate the required Fabric security model, labels, endorsement, and access reviews before publishing data products.

Mirroring readiness also depends on supported table and column features, identity settings, tenant boundaries, and connectivity. Treat Fabric as a governed data-products platform: assign product owners, define consumer SLAs, document lineage, and test replication behavior before promising analytics availability to business teams.

Use the Microsoft Fabric Azure SQL Database mirroring limitations and Fabric adoption roadmap as design inputs.

Shortcuts complement mirroring

For non-SQL data sources — Azure Data Lake Storage, Amazon S3, Dataverse, and others — Fabric shortcuts provide zero-copy, virtualized access to data without replicating it. Combined with mirroring for SQL sources, this covers the full data estate.

CAF guidance for unified data platforms

Microsoft’s Cloud Adoption Framework outlines a four-step framework for unifying your data platform: Organizational readiness → Architecture → Governance and security baselines → Operational standards. Use this as a reference when building out the customer’s Fabric practice.

Operating Model

Fabric adoption needs more than mirrored databases. Align the platform to the Fabric adoption roadmap: executive sponsorship, data culture, content ownership, content delivery scope, a Center of Excellence, governance, mentoring, community of practice, user support, system oversight, and change management.

For dc2fabric, that means every mirrored source should have a named data owner, classification and access-review process, documented refresh or replication expectations, and a support path for data consumers. The platform team provides guardrails; business data-product owners make the data usable and trusted.

← Back to Transform Platforms · Continue to Execution →