Synapse Analytics to Fabric Warehouse
Migrate Synapse workloads to Microsoft Fabric Warehouse. Instantly translate SQL code, validate results, and generate the test suite that proves it still works.
At first glance, Azure Synapse and Microsoft
Fabric Warehouse look nearly identical
Both are cloud-native analytics platforms. Both use familiar SQL and both target large-scale analytical workloads. However, they are not interchangeable. Fabric Warehouse is not a rebranding of Synapse. It is a managed, opinionated evolution with a different execution model, SQL surface, and feature set.
Key differences that matter in
migrations
Manual rewrites introduce regressions, subtle logic changes, and performance surprises. That is exactly where SQL Tran adds value, by detecting differences early, adapting behavior automatically, and making every deviation explicit.
The purpose of automated software is to free up your resources to focus only of a few percent of the codebase that absolutely requires manual review and attention, and possibly a complete rethink.
Fabric Warehouse intentionally limits and reshapes supported SQL features. Constructs that work in Synapse may be unsupported, ignored, or behave differently in Fabric.
Fabric Warehouse reduces support for procedural patterns, temporary object lifecycles, and control flow assumptions commonly used in Synapse codebases.
Distributed execution in Fabric introduces different optimization rules and removes some assumptions Synapse users rely on, especially around intermediate results and execution order.
Even small behavioral differences in analytics pipelines can lead to silent data corruption, incorrect aggregations, or performance regressions.
How SQL Tran Bridges the
Gap
SQL Tran was designed for complex real-world Synapse codebases. It uses deep static analysis and a targeted emulation and rewrite layer to adapt Synapse SQL semantics to Fabric Warehouse’s execution model, preserving intent while aligning with Fabric’s supported feature set.
Where direct equivalents exist, SQL Tran uses them. Where they don’t, SQL Tran reconstructs behavior using Fabric-native SQL patterns. It transforms Synapse SQL into idiomatic Fabric Warehouse SQL that behaves the same, or flags where exact equivalence is not possible.
What SQL Tran Handles Automatically
SQL Tran parses and analyzes every object, building a complete semantic model.
Fabric Warehouse doesn't use or need distributions the way Synapse analytics does. This, and other unnecessary constructs, are automatically removed, and table definitions are rewritten to be optimal for Fabric's distributed architecture.
Many SQL code constructs that are valid in Synapse are not supported in Fabric. SQL Tran employs a large and growing set of emulation techniques to rewrite unsupported logic into equivalent code that runs natively in Fabric, ensuring identical behavior without the need for manual intervention.
Fabric Warehouse handles identity columns differently than Synapse. SQL Tran automatically translates identity logic to be compatible with Fabric's architecture.
Fabric Warehouse has two types of temporary tables with different pros and cons compared to Synapse. SQL Tran analyzes temp table usage patterns and rewrites them to use the optimal type in Fabric, ensuring both compatibility and performance.
Rich Logic & Control Flow
Even the most complex procedural logic is automatically translated, including loops, conditionals, and error handling.
Cursors
Fabric Warehouse doesn't support T-SQL cursors. SQL Tran materializes cursor result sets into temporary tables and rewrites loops into deterministic row-by-row logic compatible with distributed execution.
Result set limiting
Constructs like SET ROWCOUNT are transformed into equivalent TOP clauses.
Selected Emulations and
Feature Adaptations
SQL Tran includes targeted emulations and rewrites for Synapse-specific behavior when migrating to Fabric Warehouse.
Synapse data types that differ in support or behavior are rewritten to Fabric Warehouse-compatible equivalents, with explicit handling of precision, collation, and implicit casting differences.
Identity behavior is preserved where Fabric supports it and rewritten into explicit value-generation logic where required.
Synapse temporary tables and intermediate constructs are adapted to Fabric-supported alternatives, ensuring predictable lifecycle and scope.
Translation with intent
Thanks to super-fast, fully custom parsers, translation of even very large Synapse codebases is practically instant. Millions of lines of SQL can be analyzed and translated in seconds. SQL Tran does more than rewrite, it aligns Synapse logic with Fabric Warehouse’s execution guarantees, making behavior predictable in production.
Unsupported Functionality
Some Synapse constructs cannot be fully preserved and are explicitly flagged. In these cases, SQL Tran preserves structure, annotates limitations, and provides clear guidance for targeted manual review instead of failing silently.
We emulate whatever is theoretically possible and let you know what needs attention
Certain system views in Synapse Analytics do not have equivalent in Fabric Warehouse at all.
Assessment and
Reporting
Assessment is instant and detailed.
This assessment is produced by running a full translation upfront, exposing exactly what will, and will not, work before migration begins.
Know exactly how many objects of each type you have.
See how complex your objects are, and how that complexity is distributed across your database.
Feature gap is automatically detected and reported on, so you know areas that require manual intervention.
Sophisticated testing framework gives you confidence that your translated database will work as expected.
Migrations available now
These are the source-to-target combinations exposed by the current SQL Tran project wizard. Start an assessment for your database and version to see what SQL Tran can translate, emulate, test, and flag for review. Preview paths are marked explicitly.
Oracle to PostgreSQL
Translate Oracle schemas and PL/SQL into PostgreSQL, then generate tests that compare source and target behavior.
Explore the scenario →Oracle to Fabric Warehouse
Move Oracle database code toward Microsoft Fabric Warehouse with semantic translation and explicit limitation reporting.
Explore the scenario →Oracle to Fabric Lakehouse
Split an Oracle workload into Spark SQL for relational objects and Python/PySpark for procedural code in Fabric Lakehouse.
Explore the scenario →Oracle to Databricks
Split an Oracle workload into Spark SQL for data objects and Python/PySpark for procedural logic on Databricks.
Explore the scenario →Oracle to Google Cloud Spanner
Translate Oracle schemas to GoogleSQL and move procedural logic into a selectable application language for Spanner.
Explore the scenario →SQL Server to Fabric Warehouse
Translate SQL Server schema and T-SQL into the Microsoft Fabric Warehouse dialect, with target limitations surfaced during assessment.
Explore the scenario →SQL Server to Azure Synapse Analytics
Refactor SQL Server schema and T-SQL for the Azure Synapse dedicated SQL pool surface.
Explore the scenario →SQL Server to PostgreSQL
Translate SQL Server schema and T-SQL into PostgreSQL and PL/pgSQL, with differences surfaced before deployment.
Explore the scenario →Azure SQL Database to Fabric Warehouse
Modernize Azure SQL Database code for Fabric Warehouse while keeping the migration inside the Microsoft data platform.
Explore the scenario →Azure SQL Database to Azure Synapse Analytics
Refactor Azure SQL Database workloads for an Azure Synapse dedicated SQL pool.
Explore the scenario →Azure SQL Database to PostgreSQL
Translate Azure SQL Database schema and T-SQL into PostgreSQL and PL/pgSQL.
Explore the scenario →Azure Synapse Analytics to Fabric Warehouse
Move a Synapse dedicated SQL pool to Fabric Warehouse with focused compatibility analysis and T-SQL refactoring.
Explore the scenario →Get SQL Tran today!
Ready to get started?