Roadmap tracker

What's New in Fabric?

Recent Microsoft Fabric roadmap updates for the community, with filters for product, release type, status, and search.

How this data is updated

This page pulls roadmap data from the Fabric GPS JSON API (with an RSS fallback if needed). We cache and refresh this feed every 24 hours, so there can be a short delay between a Fabric blog/roadmap update being published and it appearing here. If Microsoft updates or edits an existing roadmap item, those changes are picked up on the next refresh cycle.

ProductRelease typeStatusPlanned releaseModified withinSort by

1123 updates found• Q1 Jan-Mar, Q2 Apr-Jun, Q3 Jul-Sep, Q4 Oct-DecSource: Fabric GPS JSON API

Data EngineeringPublic previewPlanned

Release date: Q1 2027

[Monitoring] Integrate and Expose Spark Application CPU/Memory Usage

This feature is designed to visualize real-time metrics for Spark applications through a dashboard that allows users to monitor CPU and memory utilization at the application level for both executors and drivers, across running and completed applications. It aligns with the Fabric SaaS offering and extends Spark vCore allocation and utilization analysis. Additional details about user scenarios are available in the discussion.

Updated 17 Oct 2026Read more

Data EngineeringGeneral availabilityPlanned

Release date: Q1 2027

[GA] Job Insights Library

The general availability of the Job Insights Library provides comprehensive visibility into Spark execution by surfacing logs and metrics across queries, jobs, stages, and tasks. Users can not only analyze execution behavior directly within the experience but also download and store driver and executor logs to Lakehouse or other destinations, enabling persistent storage, deeper analysis, and integration with downstream monitoring workflows.

Updated 17 Oct 2026Read more

Data EngineeringPublic previewPlanned

Release date: Q1 2027

Preview - Runtime 2.1 (Spark 4.2 and Delta 4.4)

Microsoft Fabric Runtime 2.1 updates the Apache Spark foundation to Spark 4.2 and Delta Lake 4.4, with Python 3.14, Scala 2.13, and Java 21. For supported workloads, the Native Execution Engine is enabled by default to provide acceleration without extra configuration, while documented opt-out and controlled fallback paths preserve compatibility.Runtime 2.1 also introduces a leaner built-in package baseline with PySpark and SparkR. Customers can add compatible optional Python and R libraries through Library Manager or a Fabric Environment, reducing package conflicts while preserving flexibility for data engineering and data science workloads. The Public Preview is targeted for January 31, 2027 and includes migration guidance, known limitations, and preview support information.

Updated 17 Oct 2026Read more

Data SciencePublic previewPlanned

Release date: Q1 2027

Feature Store

Fabric feature store is a centralized system that manages the data inputs, called features, that power machine learning models. It bridges raw data and ML models, giving teams one place to define, store, serve, and discover features instead of each team building its own pipelines.

Updated 15 Oct 2026Read more

SQL databasePublic previewPlanned

Release date: Q1 2027

Outbound Access Protection for Fabric SQL DB

Microsoft Fabric's outbound access protection blocks all outbound connections from a workspace by default and only allows approved external connections through managed private endpoints or data connection rules. It applies across workloads such as Data Engineering, Data Factory, Activator, and Data Agent, giving admins granular control over external connectivity.

Updated 15 Oct 2026Read more

Conversational AnalyticsPublic previewPlanned

Release date: Q4 2026

Data Agent monitoring - Consumer to Creator feedback loop

The primary goal is to enable creators to understand how consumers interact with their data agents and bridge the gap between the data agent creators and the consumers. This comprehensive monitoring system will enable creators to have access to the feedback provided by consumer across consumption channels such as M365 Copilot, Fabric, Org App so they can refine the data agent based on customer feedback.Our initial release will only cover the M365 Copilot and will then expand to other consumption channels.

Updated 15 Oct 2026Read more

IQGeneral availabilityPlanned

Release date: Q1 2027

Ontology item GA

General availability of the Microsoft Fabric IQ Ontology item enables teams to maintain a shared model of business entities, properties, and relationships in Fabric. Modelers can connect those definitions to supported data sources and provide business context to supported analytics and agent experiences. The release covers the ontology item and its supported capabilities; individual integrations, security features, and advanced modeling capabilities retain their own release milestones.

Updated 14 Oct 2026Read more

Data FactoryPublic previewPlanned

Release date: Q1 2027

Dataflows - Merge/Upsert Support for Output Destinations

A commonly requested new capability for Output Destinations is the ability to merge, or upsert, data into previously loaded rows in the destination table. We aim to provide this support for Fabric Lakehouse destination.

Updated 13 Oct 2026Read more

SQL databaseGeneral availabilityShipped

Release date: Q3 2026

Vector Index & Search

# Generally Available: Vector Search and Vector Indexes in Azure SQL Vector Search and Vector Indexes are now generally available in Azure SQL Database. Using the [CREATE VECTOR INDEX](https://learn.microsoft.com/sql/t-sql/statements/create-vector-index-transact-sql?view=sql-server-ver17) and [VECTOR_SEARCH](https://learn.microsoft.com/sql/t-sql/functions/vector-search-transact-sql?view=sql-server-ver17) T-SQL capabilities, you can perform efficient approximate nearest neighbor (ANN) searches for AI scenarios such as semantic search, retrieval-augmented generation (RAG), recommendation systems, and image search. Built on the DiskANN algorithm, Vector Indexes enable efficient vector similarity search over large datasets while allowing you to combine vector search with relational queries and filtering directly in Azure SQL Database.By bringing AI retrieval directly into Azure SQL Database, you can build AI-powered applications on your enterprise data without requiring a separate vector database, simplifying application architecture while leveraging the reliability, security, and governance capabilities of Azure SQL Database.**Learn more:** [Vector Search & Vector Indexes documentation](https://learn.microsoft.com/sql/sql-server/ai/vectors?view=sql-server-ver17)

Updated 12 Oct 2026Read more

OneLakeGeneral availabilityPlanned

Release date: Q4 2026

Databricks - Onelake as native storage - GA / Production-ready

Production-ready support for Azure Databricks to store data in OneLake with OneLake Security enabled

Updated 9 Oct 2026Read more

Data WarehousePublic previewShipped

Release date: Q4 2026

T-SQL language improvements

New capabilities focused on making T-SQL more powerful for analytics, simpler to write, and more intuitive for data exploration.<br/>* **Advanced Analytics** Expanded statistical capabilities with functions such as `MEDIAN`, `QUANTILE`, `PERCENTILE`, `APPROX_MEDIAN`, and `APPROX_QUANTILE` for distribution analysis and large-scale analytical workloads.* **Simplified Query Syntax** New language constructs including `QUALIFY`, `GROUP BY ALL`, and `ORDER BY ALL` help reduce query complexity and improve readability of analytical queries.* **More Intuitive Querying** Developer-friendly syntax enhancements designed to make common query patterns easier to express, improving productivity and reducing the amount of code required to answer business questions.

Updated 8 Oct 2026Read more

Data WarehouseGeneral availabilityShipped

Release date: Q3 2026

VARCHAR(MAX) support in SQL analytics endpoint of Lakehouses (new metadata sync)

SQL analytics endpoints for Lakehouses will support large string and binary data using VARCHAR(MAX) and VARBINARY(MAX) types, up to 16 MB, with the new Metadata Sync feature.

Updated 8 Oct 2026Read more

Conversational AnalyticsGeneral availabilityShipped

Release date: Q3 2026

Data Agent Migration to Responses API

The OpenAI Assistants API that powers the orchestration layer for the Microsoft Fabric data agent is currently scheduled to be shut down by OpenAI on August 26, 2026. After that date, direct calls to the Assistants API will stop working. We are migrating the Data Agent UX and Programmatic experiences to the Responses API

Updated 8 Oct 2026Read more

Conversational AnalyticsPublic previewShipped

Release date: Q3 2026

Explore Large Schemas in the Data Agent Schema Explorer

Production Datawarehouses, Lakehouses, Semantic Models, and SQL DBs contain hundreds to tens of thousands of Tables, Views, Functions, and Measures. This feature highlights the requirements to support production scale database systems and semantic models across data agent scenarios within the Data Agent UX seamlessly: - Enabling the UX to allow users to load large schemas into Data Agent and select Objects easily and efficiently.- Enabling the UX to search for elements allowing users to easily identify key entities required for data agent input - Enabling pagination in the UX to seamlessly browse the schema in the Object explorer.

Updated 8 Oct 2026Read more

Conversational AnalyticsPublic previewShipped

Release date: Q3 2026

Data Agent to Upgrade Ontology Integration

Microsoft Fabric Data Agent is improving its Ontology experience so customers can build agents that deliver more accurate, reliable, and verifiable answers grounded in their business concepts and relationships. With this release, customers should see better support for complex questions, clearer visibility into answer generation, stronger customization and scoping, and improved confidence as ontologies evolve over time.

Updated 8 Oct 2026Read more

Data WarehousePublic previewPlanned

Release date: Q1 2027

Shortcuts in Fabric Data Warehouse (Public Preview)

Shortcuts in Fabric DW allow users to directly access and query data that is present in internal and external data sources, without loading them into Fabric DW. Users will have the capability to create Table Shortcuts via TSQL, UX and API

Updated 7 Oct 2026Read more

Data WarehouseGeneral availabilityPlanned

Release date: Q1 2027

ALTER COLUMN (GA)

ALTER COLUMN to non-nullable lets customers enforce "this column must always have a value" directly, in place, no more full table rebuilds just to tighten a constraint. It saves time, avoids error-prone workarounds (CTAS, drop/rename), and preserves table history. Combined with default-constraint support, existing rows get backfilled automatically, so the change is safe and non-breaking.

Updated 7 Oct 2026Read more

Data FactoryPublic previewPlanned

Release date: Q1 2027

Dataflows - New Destination: Google Cloud Storage

We are introducing Google Cloud Storage (GCS) as a new data destination for Dataflow Gen2 in Preview, enabling customers to land transformed data from Microsoft Fabric directly into Google Cloud Storage using Dataflow Gen2's low-code Power Query experience.This preview expands Dataflow Gen2's destination ecosystem to better support multi-cloud data architectures, giving customers with existing investments in Google Cloud a simple way to integrate Fabric-based transformations into their broader data estate.Key benefits and scenarios:* Publish curated outputs from Dataflow Gen2 directly to Google Cloud Storage buckets* Support multi-cloud ingestion and data sharing scenarios while centralizing transformation logic in Fabric* Enable teams to prepare and standardize data in Fabric before making it available to GCP-based analytics, processing, or downstream pipelinesDuring Preview, the Google Cloud Storage data destination is intended for evaluation and feedback, allowing customers to validate connectivity patterns, performance characteristics, and integration workflows ahead of broader production use.This release is part of our broader effort to make Dataflow Gen2 a flexible, low-code transformation layer across clouds, with ongoing investments planned to further mature and expand multi-cloud destination support in Fabric.

Updated 7 Oct 2026Read more

Data FactoryPublic previewPlanned

Release date: Q4 2026

Dataflows - New Destination: PowerPoint Presentations (PPTX)

We plan to introduce PowerPoint Presentations (PPTX) as a new data destination for Dataflow Gen2, enabling customers to turn data prepared in Microsoft Fabric into presentation files using the low-code Power Query experience.Designed for citizen users, this destination brings prepared data into a familiar tool for communicating business results. It helps teams move from data transformation to stakeholder presentations, making it easier to share results through the PowerPoint documents they already use. Key benefits and scenarios:* Create data-driven presentations: Generate PowerPoint files from Dataflow Gen2 results for business reviews, operational updates, and team briefings.* Reuse preparation logic: Build presentation outputs on existing queries and transformations rather than preparing the same data separately.* Reduce manual handoffs: Simplify the process of moving prepared data into presentations.* Work in a familiar format: Deliver PowerPoint documents that business users can open, review, and share with colleagues.During Preview, customers can evaluate PowerPoint document generation in their workflows and share feedback to help refine the experience.

Updated 7 Oct 2026Read more

Data FactoryPublic previewPlanned

Release date: Q4 2026

Dataflows - New Destination: HTML Document

We plan to introduce HTML Document as a new data destination for Dataflow Gen2, enabling customers to turn data prepared in Microsoft Fabric into shareable, browser-friendly documents using the low-code Power Query experience. Designed for citizen users, this destination meets business teams where they are today--with content they can open, read, and share in familiar tools. It bridges the gap between preparing data and communicating results, without requiring a separate reporting solution to make those results accessible.Key benefits and scenarios:* Make prepared data accessible: Deliver dataflow results as HTML documents that colleagues and stakeholders can view in a web browser.* Simplify everyday communication: Create data-driven business summaries and operational updates for teams that work primarily with documents.* Reduce repetitive work: Generate document outputs when the dataflow runs, rather than manually copying query results into separate documents. During Preview, customers can evaluate HTML document generation for their everyday workflows and provide feedback to help refine the experience.

Updated 7 Oct 2026Read more

Data FactoryGeneral availabilityPlanned

Release date: Q4 2026

Dataflows - New Destination: PostgreSQL

The PostgreSQL data destination in Dataflow Gen2 will become generally available, enabling customers to prepare and transform data in Microsoft Fabric and publish the results directly into PostgreSQL tables using the low-code Power Query experience. This release helps customers connect their Fabric data preparation workflows with existing PostgreSQL investments, making transformed data available for reporting, analytics, and downstream applications.Key benefits and scenarios:* Deliver curated data to PostgreSQL: Publish cleaned and standardized outputs directly into PostgreSQL tables. * Simplify recurring data delivery: Use dataflow refreshes to load transformed results, reducing the need for separate export-and-load jobs.* Complement existing read connectivity: Combine PostgreSQL source and destination support to prepare data in Fabric and deliver the results back to PostgreSQL. General Availability will make this destination an option for customers looking to bring their Fabric-to-PostgreSQL data preparation and loading workflows into production.

Updated 7 Oct 2026Read more

Data FactoryPublic previewPlanned

Release date: Q4 2026

Dataflows - Send Email with HTML Formatting

We are enhancing the previously released Send Email action in Dataflow Gen2 with rich HTML formatting, enabling customers to create more readable, data-driven emails with query results--such as tables--embedded directly in the message body.This enhancement meets business users where they already work: their inbox. Customers can combine formatted content with data-bound tables to communicate results clearly, without manually copying data into emails or requiring recipients to open a separate report to see the included results.Key benefits and scenarios:* Embed data in context: Include data-bound tables alongside explanatory text, keeping results and their business context together.* Improve readability: Use rich HTML formatting to create structured, easy-to-scan email updates.* Simplify business communication: Deliver data-driven operational summaries and status updates through the existing Send Email action.* Reduce manual preparation: Populate email content from query results rather than repeatedly assembling messages by hand.This enhancement will be available in Preview, giving customers an opportunity to try rich HTML email outputs and share feedback to help refine the experience.

Updated 7 Oct 2026Read more

Data FactoryGeneral availabilityShipped

Release date: Q3 2026

Dataflows - Dataflow Diagnostics Download

We plan to make Diagnostics Download in Dataflow Gen2 generally available, providing a production-ready way to capture and share detailed runtime diagnostics for dataflow executions.Moving this feature to GA means it is ready for production usage, with consistent behavior, reliability, and support suitable for business-critical data pipelines.Key benefits and scenarios:* Download run-level diagnostic packages including execution details and runtime metadata* Speed up troubleshooting, root-cause analysis, and performance investigations* Enable easier collaboration between data teams, administrators, and Microsoft support using standardized diagnostic artifactsBy taking Diagnostics Download to GA, we're strengthening observability and supportability for Dataflow Gen2, making it easier to operate, troubleshoot, and trust dataflows at production scale.

Updated 7 Oct 2026Read more

Data FactoryPublic previewShipped

Release date: Q3 2026

Dataflows - Shared Queries

Shared Queries in Dataflow Gen2 will be available in Preview, enabling users to save commonly used Power Query transformations within a Fabric workspace, share and reuse them across multiple dataflows and collaborators.This preview extends the reuse-first authoring model to teams, making it easy to apply trusted, reusable logic--such as data cleansing steps, business rules, or standard joins--consistently across projects without rebuilding queries from scratch.Key benefits and scenarios:* Share common transformation logic across workspace members to accelerate team-wide development* Promote consistency and standardization of data preparation patterns across dataflows* Reduce duplication and improve maintainability by centralizing reusable query logic* Enable faster onboarding and collaboration when multiple authors work with similar source systemsDuring Preview, Shared Queries is intended for experimentation and early team adoption, helping organizations validate collaborative transformation patterns and provide feedback as we continue evolving the experience.

Updated 7 Oct 2026Read more

Data sourced from Fabric GPS via JSON API and RSS endpoints.