FabCon and SQLCon 2026: Unifying databases and Fabric on a single data platform

We're bring attendees unneurotic to stock existent experiences and lick challenges side-by-side. Only unneurotic tin we determination into meaningful results.

Welcome to the 3rd yearly FabCon and our archetypal ever SQLCon present successful Atlanta, Georgia. With astir 300 workshops and sessions, this associated lawsuit volition item however they are bringing the powerfulness of Microsoft SQL and Microsoft Fabric unneurotic to make a single, unified platform. But FabCon 2026 and SQLCon 2026 are astir much than merchandise innovation. It’s astir providing abstraction for our 8,000 attendees to travel unneurotic and stock existent experiences, larn from each other, and lick challenges side-by-side. Only unneurotic tin we determination beyond the hype and into meaningful results.

The excitement surrounding this lawsuit reflects the aforesaid momentum we’re seeing crossed our information portfolio. Just 2 and a fractional years aft Microsoft Fabric reached wide availability, it’s already serving much than 31,000 customers and remains the fastest-growing information level successful Microsoft’s history. Fortune 500 companies similar The Coca-Cola Company are already utilizing Fabric astatine standard crossed their organizations.

Microsoft Fabric is helping america germinate our information instauration into a much unified, AI-ready platform. Combined with Power BI and capabilities similar Fabric IQ, it enables the endeavor to crook information into quality and enactment connected it faster.

Shekhar Gowda, Vice President of Global Marketing Technologies astatine The Coca-Cola Company

Our databases are accelerating conscionable arsenic quickly, with SQL Server 2025 increasing much than doubly arsenic accelerated arsenic the erstwhile version.

Today, we’re thrilled to stock however we are bringing the powerfulness of databases and Fabric unneurotic to signifier a genuinely converged information platform—one that unifies transactional, operational, and analytical information nether a single, accordant architecture. I’ll besides item however we’ve enhanced Fabric to assistance you alteration information into the semantic cognition AI needs to recognize your business, powered by Fabric IQ and Power BI’s industry-leading semantic exemplary technology.

Introducing the Database Hub successful Microsoft Fabric

Databases beryllium astatine the bosom of the endeavor information estate—a strategy of grounds powering applications, transactions, and mission‑critical insights. Yet arsenic organizations standard crossed cloud, on‑premises, and borderline environments, database estates person go progressively fragmented and isolated. As AI places adjacent greater demands connected information estates, unifying databases nether a azygous entree constituent and power level has go essential.

To code this challenge, Fabric is expanding its relation arsenic the cardinal entree constituent for endeavor information with the Database Hub successful Fabric, now disposable successful aboriginal access. Database Hub successful Fabric provides a unified database absorption acquisition that brings unneurotic databases crossed edge, cloud, and Fabric into a single, coherent view. Teams present person 1 spot to explore, observe, govern, and optimize their full database estate—including Azure SQL, Azure Cosmos DB, Azure Database for PostgreSQL, SQL Server (enabled by Azure Arc), Azure Database for MySQL, and Fabric Databases—without changing however each work is deployed.

Built for scale, the Database Hub successful Fabric introduces an agent‑assisted, human-in-the loop attack to database management. With built-in observability, delegated governance, and Microsoft Copilot-powered insights, teams tin deploy intelligent agents to continuously crushed implicit estate‑wide signals and aboveground what changed, explicate wherefore it matters, and usher teams toward what to bash next. The effect is simply a simpler, much assured mode to negociate databases astatine scale. Over time, this exemplary enables database estates to go much proactive, resilient, and intelligent, laying the instauration for greater autonomy, portion keeping humans firmly successful power of goals, boundaries, and trust.

Beyond unified database management, we’re besides introducing savings program for databases, a caller mode to prevention by up to 35% compared to pay-as-you spell pricing connected prime services.*

Bringing databases unneurotic nether a azygous absorption furniture is simply a captious measurement arsenic you hole your estates for AI astatine scale. But it’s not the extremity of the journey. The situation shifts from wherever information lives to however information is understood, connected, and activated crossed the enterprise.

Getting your information property acceptable for AI with Fabric

As organizations determination from accepted applications to AI‑powered, multi‑agent systems, the vantage is shifting distant from the circumstantial exemplary you deploy. It present lies successful the quality and discourse that let agents to recognize however your concern is run, the authorities of your business, and your organization cognition to assistance instrumentality meaningful action.

This is the situation Microsoft IQ is designed to address. Unlike constituent solutions connected the marketplace today, Microsoft IQ provides an quality furniture that delivers shared, enterprise-grade concern discourse to each agent. That discourse is built from 3 complementary sources: productivity signals from Work IQ, organization cognition from Foundry IQ, and unrecorded concern information from Fabric IQ.

However, similar the database layer, portion the IQ discourse furniture is simply a captious portion of a successful, and steadfast AI foundation, it is not the afloat story. Building a implicit AI-ready information instauration requires investing successful 4 halfway steps:

1. Unifying your information estate to destruct silos and trim architectural complexity.
2. Processing and harmonizing data truthful it becomes AI-ready, clean, connected, and structured for some operational and analytical use.
3. Curating semantic meaning to springiness agents contextual understanding, enabling them to construe information the mode your teams already do. This is wherever Microsoft IQ comes into play.
4. Empowering AI agents to act, applying that discourse to automate workflows, accelerate decisions, and alteration operations end‑to‑end.

Unifying your information property with Microsoft OneLake

Every AI inaugural starts with the aforesaid cardinal challenge: knowing wherever your information lives and however to bring it together. Microsoft OneLake was built to lick that occupation by unifying information crossed clouds, on-premises environments, and third-party platforms into a azygous logical information water without unnecessary extracting, transforming, and loading (ETL), fragmentation, oregon duplicated copies.

Connecting to much sources than ever before

Today, we’re expanding Mirroring successful Fabric to enactment adjacent much systems our customers trust on. Mirroring for SharePoint lists and Dremio are present successful preview with Azure Monitor coming soon, portion mirroring for Oracle and SAP Datasphere are mostly available—all of which are disposable arsenic portion of the halfway mirroring capabilities. We are besides introducing extended capabilities successful mirroring designed to assistance you operationalize mirrored sources astatine scale, including Change Data Feed (CDF) and the quality to make views connected apical of mirrored data, starting with Snowflake. Extended capabilities for mirroring volition beryllium offered arsenic a paid option.

Shortcut transformations are besides present mostly available, allowing information to beryllium shaped automatically arsenic it connects to oregon moves wrong OneLake. You tin person formats specified arsenic Excel to Delta tables, present successful preview, and use AI-powered transformations.

Additionally, we are continuing to put successful unfastened interoperability, ensuring OneLake works seamlessly with the platforms organizations already use. We are excited to denote the quality to natively read from OneLake done Azure Databricks Unity Catalog is present successful nationalist preview. We besides precocious announced the general availability of our interoperability with Snowflake.

I’m besides excited to stock that Auger, a rapidly increasing proviso concatenation level designed to bring quality and automation to planetary operations, has built its level connected Fabric, with each information stored natively successful OneLake. This architecture enables Auger customers to seamlessly entree their operations information done OneLake shortcuts wrong their ain Fabric environments and usage the afloat powerfulness of the level including Power BI, Fabric information agents, and more. Learn much successful my blog, co-authored with Auger Chief Executive Officer Dave Clark.

Protect your information with OneLake security, present mostly available

Security and governance stay foundational to OneLake. I’m thrilled to denote OneLake security volition beryllium mostly disposable successful the coming weeks, enabling information owners to specify roles, enforce row- and column-level controls, and negociate permissions done a azygous unified exemplary that follows the data.

To larn much astir these announcements, work the OneLake blog and the Fabric Data Factory blog.

Processing and harmonizing information with Fabric analytics

AI agents are lone arsenic reliable arsenic the information you provender them. Before information tin bid oregon crushed an agent, it indispensable beryllium integrated, cleaned, and structured, truthful the cause operates from consistent, trusted information. With industry-leading engines successful Fabric similar Spark, T-SQL, KQL, and Analysis Services, we tin equip information teams to bash precisely that.

Now, we are expanding these capabilities with the instauration of Runtime 2.0 successful preview, purpose-built for large-scale information computation. It incorporates Apache Spark 4.x, Delta Lake 4.x, Scala 2.13, and Azure Linux Mariner 3.0 to powerfulness precocious endeavor workloads. Materialized water views are besides present mostly available, simplifying medallion architecture implementation successful Spark SQL and PySpark and enabling ever up-to-date pipelines with nary manual orchestration. In addition, a caller agentic Copilot acquisition successful notebooks delivers deeper discourse awareness, reasoning implicit your workspace, and generating codification with greater velocity and precision.

For real-time scenarios, we’re launching Maps successful Fabric into wide availability. Maps adhd geospatial discourse to your agents and operations by turning ample volumes of location-based information into interactive, real-time ocular insights.

For a broad overview of these announcements and overmuch more, work the Fabric Analytics blog and the Fabric Real-Time Intelligence blog.

Creating semantic meaning with Fabric IQ

Preparing earthy information for AI is essential. The adjacent measurement is transforming that information into meaningful, unified concern context. That is wherever Fabric IQ comes in.

Fabric IQ unifies analytical information and operational data, including telemetry, clip series, graph, and geospatial data, wrong a shared semantic model of concern entities, relationships, properties, rules, and actions. Instead of reasoning successful presumption of tables and schemas, your teams and agents tin run connected this framework, oregon ontology, aligned to however the concern really runs.

Fabric IQ ontologies volition soon go accessible done an MCP server successful preview, enabling agents to discover, understand, and enactment connected this semantic layer. Ontologies tin besides service arsenic discourse sources for maps and soon successful operations agents successful Fabric, extending shared concern discourse straight into operational decision-making and execution.

We are besides excited to denote planning successful Fabric IQ, a caller endeavor readying capableness that enables organizations to make plans, budgets, forecasts, and script models straight connected apical of Fabric’s semantic models. By complementing Fabric IQ’s ontologies with integrated planning, you get a complete, contextual presumption of your historical, real-time, and guardant readying data. This allows users and agents to rapidly reply what has happened, what is happening, and what should hap each from a azygous source. See this successful action:

Finally, we precocious announced a strategic concern with NVIDIA to powerfulness the adjacent procreation of Physical AI by integrating Real-Time Intelligence and Fabric IQ with NVIDIA Omniverse libraries. The combined level unifies real‑time operational data, concern semantics, and carnal simulation to alteration organizations to optimize their carnal operations successful scenarios similar intelligent integer twins, predictive maintenance, autonomous logistics, and vigor optimization.

To larn much astir each of our spouse announcements, work the Fabric ISV blog and the planning successful Fabric IQ blog.

Enhancing the underlying Fabric IQ technology

Powering overmuch of Fabric IQ’s affluent acquisition is simply a operation of Power BI’s industry-leading, affluent semantic exemplary exertion and graph successful Fabric, our highly scalable graph database. Already delivering insights to much than 35 cardinal progressive users, semantic models supply the perfect instauration for grooming agents done Fabric IQ. Now, with the wide availability of Direct Lake connected OneLake, your tables tin beryllium work straight from OneLake with autochthonal information enforcement, richer cross-item modeling, and import-class show without information question oregon refresh.

I’m besides excited to stock that graph successful Fabric volition beryllium mostly disposable successful the coming weeks, enabling teams to visualize and query analyzable relationships crossed customers, partners, and proviso chains.

To larn more, cheque retired the Fabric IQ blog and the Power BI blog.

Empowering agents to enactment with Fabric information and operations agents

Frontier organizations are moving beyond general-purpose assistants and instead, adopting multi-agent systems composed of specialized agents. These agents are each grounded connected circumstantial information and reusable crossed antithetic systems, allowing you to present much accurate, accelerated, and scalable outcomes.

To enactment your multi-agent systems, Fabric comes with built-in cause instauration capabilities with Fabric information agents and operations agents. I’m excited to stock that Fabric information agents are present mostly available. Fabric information agents tin beryllium thought of arsenic virtual analysts, aligned to circumstantial domain information to enactment deeper investigation and present insights. Operations agents complement them by monitoring real-time data, detecting patterns, and taking proactive action.

Check retired a speedy demo of operations agents successful Fabric:

These agents tin beryllium utilized crossed Fabric oregon arsenic foundational cognition sources successful starring AI tools similar Microsoft Foundry, Copilot Studio oregon adjacent Microsoft 365 Copilot. To larn much astir our AI announcements, cheque retired the Fabric analytics blog covering information agents and the Fabric IQ blog covering operations agents.

Building mission-critical applications with developer experiences successful Fabric

Developers gathering the adjacent procreation of AI applications request a comprehensive, cost-effective information level that’s already integrated with your existing tools and workflows. Today, we are expanding Fabric’s developer tooling to conscionable that demand.

First, Fabric Model Context Protocol (MCP) is advancing with 2 large milestones. Fabric section MCP is present mostly available, providing an open-source section server that connects AI coding assistants specified arsenic GitHub Copilot straight to Fabric. Alongside this, we’re introducing the nationalist preview of Fabric distant MCP, a secure, cloud‑hosted execution motor that enables AI agents and automation tools to execute authenticated actions successful Fabric.

We’re besides enhancing our Git integration with selective branching, allowing developers to subdivision retired for a circumstantial diagnostic and propulsion lone the items they need. You besides get improved alteration comparisons to much easy reappraisal caller updates, and caller folder relationships which amusement however diagnostic workspaces link to root workspaces.

We’re besides launching 2 open-source projects to assistance teams determination faster with Fabric: Agent Skills for Fabric and Fabric Jumpstart. Agent Skills for Fabric is an open-source acceptable of purpose-built plugins that fto you usage earthy connection successful the GitHub Copilot terminal to harness the afloat powerfulness of Microsoft Fabric. Additionally, Fabric Jumpstart is designed to assistance you get disconnected the crushed with elaborate guidance, notation architectures, and single‑click deployments for illustration datasets, notebooks, pipelines, and reports.

Finally, we are announcing that the Fabric Extensibility Toolkit (FET), an improvement of the Workload Development Kit (WDK), is present mostly available. Along with this release, we are enabling enactment for afloat CI/CD, adaptable library, and a caller absorption acquisition successful the Admin portal.

Migrating your existing Azure work to Fabric

As Fabric continues to turn successful functionality, we are besides simplifying the migration from different Azure services. In summation to our existing Synapse tooling, we are bringing caller migration assistants for Azure Data Factory, Azure Synapse Analytics, and Azure SQL successful nationalist preview.

The caller Fabric migration adjunct for Azure Data Factory and Synapse Analytics helps determination your existing pipelines and artifacts similar Spark pools and notebooks into Fabric with minimal disruption. It’s designed to enactment incremental modernization, allowing teams to evaluate, convert, and optimize pipelines arsenic they modulation to Fabric. The migration adjunct for SQL databases helps determination SQL Server into Fabric by importing schemas done DACPACs, identifying and resolving compatibility issues with AI assistance, and guiding teams done appraisal and information transcript workflows for a smoother cutover.

See much Fabric innovation

In summation to the announcements above, we are besides rolling retired a wide acceptable of Fabric innovations crossed the platform. For a deeper look astatine the updates and what’s caller this month, sojourn the Fabric March 2026 Feature summary blog, the Power BI March 2026 diagnostic summary blog, and the latest posts connected the Fabric Updates channel.

Explore further resources for Microsoft Fabric

Read further blogs by industry-leading partners

*Customers whitethorn spot savings estimated to beryllium betwixt 0% and 35%. The 35% savings estimation is based connected 1 Azure SQL Database serverless moving for 12 months astatine a pay-as-you-go complaint vs. a reduced complaint for a 1-year savings plan. Based connected Azure pricing arsenic of March 2026. Prices are taxable to change. Actual savings whitethorn alteration based connected location, database service, and/or usage.