What’s new in Fabric Warehouse – August 2025 Recap
Introduction
Welcome to the August 2025 edition of What’s New in Fabric Warehouse. As summer winds down, despite August being a slower month, our team continued to deliver meaningful updates. We shipped several new features focused on enhancing data ingestion, improving the data management, and streamlining security. At the same time, much of our energy is going into preparing exciting announcements for FabCon Vienna — stay tuned for what’s coming next. Whether you’re optimizing workloads, building with SQL, or exploring new integrations, this roundup highlights improvements we think you’ll find valuable.
Quick Roundup of What’s New
Ingest, Load, and Sync Data
Get data into Fabric from multiple sources with flexibility, automation, and governance.
- JSON Lines Support in OPENROWSET for Fabric Data Warehouse and Lakehouse SQL Analytics Endpoint (Preview)
JSON Lines (JSONL) support in the OPENROWSET(BULK) function for Microsoft Fabric Data Warehouse and SQL Analytics Endpoints for Lakehouses is now available in preview. The OPENROWSET(BULK) function allows you to query external data stored in the lake using well-known T-SQL syntax. With this update, you can now also query files in JSON Lines format, expanding the range of supported formats and simplifying access to semi-structured data. - Improvements in Copy job
Check out these updates from Copy job that offer more granular control over data ingestion workflows in Fabric Warehouse. With features like Reset Incremental Copy and JSON Format Support, teams can streamline ingestion from semi-structured sources and simplify ETL setup. The new Copy job activity in pipelines (Preview) also brings visibility and orchestration flexibility, making it easier to manage Warehouse adjacent workflows. These enhancements reflect Microsoft Fabric’s ongoing commitment to simplifying data ingestion and boost performance for Warehouse users. - Load data from network-protected Azure Storage accounts to Microsoft OneLake with AzCopy
AzCopy is a powerful, high-performance tool for moving data between Azure Storage and Microsoft OneLake, and it’s the preferred choice for large-scale transfers thanks to its simplicity and built-in optimizations. With recent updates, AzCopy now supports copying data from firewall-enabled Azure Storage accounts into OneLake using trusted workspace access. For Fabric Warehouse users, this removes a common networking roadblock and makes downstream ingestion with COPY INTO simpler and more secure.
Query, Transform, and Optimize for Insights
Write performant, reusable SQL and run queries fast – at scale.
- SHOWPLAN_XML set statement (Generally Available)
SHOWPLAN_XML is now Generally Available for the Fabric Data Warehouse and Lakehouse SQL analytics endpoint! The SHOWPLAN_XML set statement and its ‘Display estimated execution plan’ counterpart in SQL Server Management Studio are staple tools for investigating the details of a T-SQL queryplan: providing information on planned data movements, resource estimates, operator choices, and more. - OBJECT_SCHEMA_NAME TSQL
Introducing theOBJECT_SCHEMA_NAMEmetadata scalar function in Fabric Warehouse to align with SQL Server behavior. This function takes an object ID and returns the schema name that owns the object—just like in SQL Server. For warehouse users, this means smoother migration of existing SQL scripts, easier compatibility across environments, and more consistent metadata operations when managing and analyzing objects. - Introducing the Item History Page in Microsoft Fabric Capacity Metrics App (Preview)
The new Item History page brings 30 days of compute usage to life with interactive visuals and slicers. You can quickly spot overall consumption trends, then drill into item-level details to see exactly which Warehouses or SQL analytics endpoints drive the highest CU usage. This makes it easier for capacity admins to identify heavy consumers, optimize workloads, and better forecast future needs.
Secure, Govern, and Manage with Confidence
Ensure compliance, governance, and transparency for your data and models.
- Fabric workspace-level Private Link (Preview)
In May 2024, we announced the General Availability of Private Link for Fabric Tenants. Since then, we’ve heard feedback requesting more granular network security controls. Private link for Fabric workspaces empowers your organization to secure your Fabric workspaces with fine-grained network isolation, allowing private, secure access from your virtual network — without traversing the public internet. With this update, customers can strengthen their security posture by isolating workspaces to trusted network perimeters and ensuring that inbound connections to the warehouse are only allowed from approved locations. - Sunsetting Default Semantic Models – Microsoft Fabric
Based on your feedback, Fabric is retiring auto-generated semantic models in favor of giving you full control with custom semantic models—the ones you already know and trust. This change enhances transparency, improves governance, and provides greater flexibility in managing the lifecycle of your data assets. - Streamlining Data Management with Workspace Collation Settings in Microsoft Fabric Warehouse and SQL Endpoint
We’ve introduced workspace-level collation settings in Fabric Warehouse and SQL analytics endpoints to make migrations smoother and ensure consistency. Admins can now set a default collation across the workspace while still overriding it for specific items. Soon, item-level enhancements will provide even finer control with clearer options, snapshot consistency, and better shortcut management. Motherson, a global automotive supplier, has already seen the benefits – “The change meant we could run most of our SQL Server scripts in Fabric Data Warehouse without significant rewrites, reducing friction in code conversion, validation, and deployment“.
Enable AI, Copilot & Agentic Workflows
Enable Copilot and AI to reason over data in Warehouse to deliver insights throughout your organization.
- Data Warehouse: Copilot & Data Agent
Fabric Warehouse now has an entry point to directly create data agents from warehouse in the SQL query editor! While Copilot provides real-time, interactive support for building schemas, generating SQL, and exploring data, Data Agent enables customizable, natural language-driven querying tailored to organizational needs. Together, they streamline workflows, improve collaboration between technical and non-technical users, and reduce errors by combining Copilot’s guidance with Data Agent’s context-aware query generation. - Fabric Data Agents + Microsoft Copilot Studio: A New Era of Multi-Agent Orchestration
This preview introduces multi-agent orchestration by connecting Fabric Data Agents with Copilot Studio. Agents can now collaborate using Model Context Protocols (MCP), enabling richer, more complete responses across your warehouse or SQL analytics endpoint data. - Empowering Agentic AI by Integrating Fabric with Azure AI Foundry
Fabric Data Agents now integrate with Azure AI Foundry, enabling organizations to build intelligent, domain-specific AI agents grounded in enterprise data. These agents can query across warehouses, sql endpoints, and semantic models to deliver contextual insights.
Build, Develop & Deploy with Modern Tooling
Empower developers and data engineers with CLI, IDE, and CI/CD tools for Fabric.
- NuGet Gallery | Microsoft.SqlPackage 170.1.61
Fabric Warehouse now supports Scalar Functions in database projects, bringing more flexibility to development and deployment. Warehouse developers using SDK-style projects can now extract and import scalar functions directly from VS Code, Database Projects, or SQLPackage.exe. This extends CI/CD scenarios with familiar client tools, streamlining development workflows and making it easier to manage functions alongside the rest of your warehouse code. - Unlocking Flexibility in Fabric: Introducing Multiple Scheduler and CI/CD Support
Fabric now supports multiple schedulers per T-SQL notebook, allowing different job types and trigger intervals (e.g., daily query runs and weekly publish tasks) on the same notebook without duplication. With built-in CI/CD integration, scheduler configurations are versioned, deployed, and monitored alongside the notebook, ensuring scalable, reliable, and auditable automation.
Docs Updates
August’s key documentation updates:
- Performance Guidelines in Fabric Data Warehouse
Boost the performance of your Fabric warehouse workloads with proven best practices. From faster ingestion and smarter table design to optimized queries and statistics, these strategies help your Warehouse and SQL analytics endpoints run faster, scale better, and deliver more impact.
- What’s New in Fabric
Continuously updated with the latest features, fixes, and enhancements across Fabric.
- Security Fundamentals
Updated to reflect new features like workspace level private link.
Next Steps
We’re keeping a few surprises under wraps—but you won’t want to miss what’s coming next. Join us at FabCon and stay tuned to the blog for first looks at the latest innovations in Fabric Warehouse and feature announcements.
Got a feature idea? Submit it at https://aka.ms/fabricideas
