Microsoft Fabric Updates Blog

Welcome to Fabric Data Warehouse 

Fabric Data Warehouse is a relatively recent addition to the market, designed with a next-generation architecture and developed from the ground up in the 2020s. Drawing on decades of database and cloud innovation, Fabric Data Warehouse delivers advanced capabilities for modern organizations. Following FabCon Europe, we wish to highlight notable customer success stories and introduce key features (such as MERGE, Data Agent, and Customer Managed Keys) announced during the conference. 


Built for Speed

Fabric Data Warehouse leverages a lakehouse architecture, is fully serverless, instantly scales to hundreds of petabytes, and is seamlessly integrated into the Fabric data platform. The result, a data warehouse that’s ultra-modern, lightning-fast, future-proof, and it’s only getting better. 

This flexible architecture is especially amenable to iteration. In relentless pursuit of industry-best price-performance, we worked with customers to deliver over 40 significant performance improvements so far in 2025, resulting in a 36% performance boost in typical industry benchmarks. 

Customers are feeling the compounding positive effects of these investments in performance of their own real-world workloads. In case of Willis Tower Watson, a highly complex query that used to process 80 Terabytes in 64 hours on Azure Synapse now only needs 29 hours to process over 400 Terabytes!


“Using Microsoft Fabric F2048, a 47% performance improvement was achieved while processing 5x more data compared to the previous Synapse DWU-30,000 environment. At scale, Fabric successfully processed 464 TB of data in a single query.”

– Nigel Davis, Technology Director, Willis Tower Watson

In the coming weeks stay tuned for even more improvements, such as incremental stats refresh, proactive stats refresh, and clustering.

Focus on the Enterprise 

Fabric Data Warehouse is designed to meet the needs of enterprise customers, with features that enhance security, compliance, and manageability. For example, Fabric Data Warehouse is architected to seamlessly fail over to alternative zones in case of zone failure, and queries are resilient to individual node failure.  

Fabric Data Warehousing – built for the modern, innovative enterprise – and enabling anyone within an organization – from developers, analysts, to business professionals.


Recently announced Enterprise-focused features:

  • PrivateLink enables you to secure network access to your Fabric environments. 
  • Customer Managed Encryption Keys, secure your data with your own keys. 
  • Outbound Access Protection help you limit network access to approved endpoints. 
  • Highly requested varchar(max) and varbinary(max) data types offer the flexibility of storing and analyzing larger string values. 
  • MERGE operation combines INSERT, UPDATE, and DELETE into one conditional SQL statement, empowering you to easily clean and transform your data. 
  • Audit logs are now easier to inspect and analyze in the UI.
  • You can now query JSON files in OneLake and ADLS using OPENROWSET(JSONL). 
  • Workspace collation settings enable you to configure your data warehouse for case sensitivity. 
  • Finally, the highly requested IDENTITY columns will be made available in the coming weeks. 

Developer and Migration Tools 

Fabric Data Warehouse continues to expand its capabilities for developers and data professionals:

  • Fabric Data Agent is now available in the Fabric Data Warehouse.
  • Data-driven alerts enable you to set alerts based on query results. 
  • Fabric Data Warehouse now integrates with SSMS. 
  • Copilot is getting smarter – it can now better optimize your SQL, it now integrates with documentation, it supports multi-turn chat and features a number of improvements in SQL generation! 
New developer experiences – default workspace collations, Copilot enhancements, and data-driven alerts.

At last, to help you move to Fabric Data Warehouse, we invested in migration tooling.

The Migration Assistant provides the industry’s only AI powered, self-service warehouse migration experience

Today, our intelligent Warehouse Migration Assistant is now Generally Available (GA) and folks are already using it to move their organizations to Fabric.


“Migrating to Microsoft Fabric Data Warehouse was seamless thanks to the Fabric Migration Assistant. We transitioned without disrupting delivery, and post-migration, we’ve quadrupled our environments to boost development throughout — while cutting costs by 50% compared to legacy Azure SQ L dedicated pools. Fabric’s support for T-SQL and stored procedures made adoption smooth for our teams.”

– Denis Zorenko, Principal Product Manager, Kantar

What’s Next?

We encourage you to check out Fabric Data Warehouse and join the ever-growing community of over 20,000 warehouse customers and hundreds of thousands of developers and data practitioners.

  • Check out our public roadmap and follow along—we’re shipping daily and we’re just getting started. 
  • Submit a new feature request via Ideas Portal.

Related blog posts

Welcome to Fabric Data Warehouse 

February 17, 2026 by Virginia Roman

We’re introducing billing reporting updates that make it easier to track AI-related usage in Microsoft Fabric. New AI Functions operation Until now, Fabric AI functions usage was reported under other operations, such as Spark-related operations, or Dataflows Gen2-related operations, depending on where the functions were used. To provide more transparency, Fabric AI functions will have … Continue reading “Billing updates: new operations for Fabric AI functions and AI services”

February 3, 2026 by Arun Ulagaratchagan

Data teams today are under extraordinary pressure. Expectations around analytics and AI have never been higher, yet enterprise data continues to live across a patchwork of systems, tools, and platforms. The result is friction, duplication, and complexity, making it harder for data teams to provide a unified, real-time view of their business. Microsoft and Snowflake … Continue reading “Microsoft OneLake and Snowflake interoperability (Generally Available)”