FabCon and SQLCon 2026: Unifying databases and Material on a single information platform

We’re convey attendees collectively to share actual experiences and remedy challenges side-by-side. Solely collectively can we transfer into significant outcomes.

Welcome to the third annual FabCon and our first ever SQLCon right here in Atlanta, Georgia. With practically 300 workshops and periods, this joint occasion will spotlight how they’re bringing the facility of Microsoft SQL and Microsoft Material collectively to create a single, unified platform. However FabCon 2026 and SQLCon 2026 are about greater than product innovation. It’s about offering area for our 8,000 attendees to come back collectively and share actual experiences, study from one another, and remedy challenges side-by-side. Solely collectively can we transfer past the hype and into significant outcomes.

The joy surrounding this occasion displays the identical momentum we’re seeing throughout our information portfolio. Simply two and a half years after Microsoft Material reached normal availability, it’s already serving greater than 31,000 prospects and stays the fastest-growing information platform in Microsoft’s historical past. Fortune 500 corporations like The Coca-Cola Firm are already utilizing Material at scale throughout their organizations.

Microsoft Material helps us evolve our information basis right into a extra unified, AI-ready platform. Mixed with Energy BI and capabilities like Material IQ, it allows the enterprise to show information into intelligence and act on it sooner.

Shekhar Gowda, Vice President of International Advertising and marketing Applied sciences at The Coca-Cola Firm

Our databases are accelerating simply as shortly, with SQL Server 2025 rising greater than twice as quick because the earlier model.

Right this moment, we’re thrilled to share how we’re bringing the facility of databases and Material collectively to kind a really converged information platform—one which unifies transactional, operational, and analytical information underneath a single, constant structure. I’ll additionally spotlight how we’ve enhanced Material that will help you remodel information into the semantic data AI wants to know your enterprise, powered by Material IQ and Energy BI’s industry-leading semantic mannequin expertise.

Introducing the Database Hub in Microsoft Material

Databases sit on the coronary heart of the enterprise information property—a system of file powering purposes, transactions, and mission‑important insights. But as organizations scale throughout cloud, on‑premises, and edge environments, database estates have change into more and more fragmented and remoted. As AI locations even larger calls for on information estates, unifying databases underneath a single entry level and management aircraft has change into important.

To handle this problem, Material is increasing its position because the central entry level for enterprise information with the Database Hub in Material, now accessible in early entry. Database Hub in Material supplies a unified database administration expertise that brings collectively databases throughout edge, cloud, and Material right into a single, coherent view. Groups now have one place to discover, observe, govern, and optimize their whole database property—together with Azure SQL, Azure Cosmos DB, Azure Database for PostgreSQL, SQL Server (enabled by Azure Arc), Azure Database for MySQL, and Material Databases—with out altering how every service is deployed.

Constructed for scale, the Database Hub in Material introduces an agent‑assisted, human-in-the loop method to database administration. With built-in observability, delegated governance, and Microsoft Copilot-powered insights, groups can deploy clever brokers to repeatedly purpose over property‑huge alerts and floor what modified, clarify why it issues, and information groups towards what to do subsequent. The result’s an easier, extra assured approach to handle databases at scale. Over time, this mannequin allows database estates to change into extra proactive, resilient, and clever, laying the inspiration for larger autonomy, whereas maintaining people firmly accountable for targets, boundaries, and belief.

Past unified database administration, we’re additionally introducing financial savings plan for databases, a brand new approach to save by as much as 35% in comparison with pay-as-you go pricing on choose providers.*

Bringing databases collectively underneath a single administration layer is a important step as you put together your estates for AI at scale. However it’s not the top of the journey. The problem shifts from the place information lives to how information is known, related, and activated throughout the enterprise.

Getting your information property prepared for AI with Material

As organizations transfer from conventional purposes to AI‑powered, multi‑agent methods, the benefit is shifting away from the precise mannequin you deploy. It now lies within the intelligence and context that permit brokers to know how your enterprise is run, the state of your enterprise, and your institutional data to assist take significant motion.

That is the problem Microsoft IQ is designed to handle. In contrast to level options in the marketplace as we speak, Microsoft IQ supplies an intelligence layer that delivers shared, enterprise-grade enterprise context to each agent. That context is constructed from three complementary sources: productiveness alerts from Work IQ, institutional data from Foundry IQ, and reside enterprise information from Material IQ.

Nevertheless, just like the database layer, whereas the IQ context layer is a important a part of a profitable, and wholesome AI basis, it’s not the complete story. Constructing a whole AI-ready information basis requires investing in 4 core steps:

1. Unifying your information property to get rid of silos and cut back architectural complexity.
2. Processing and harmonizing information so it turns into AI-ready, clear, related, and structured for each operational and analytical use.
3. Curating semantic that means to offer brokers contextual understanding, enabling them to interpret information the best way your groups already do. That is the place Microsoft IQ comes into play.
4. Empowering AI brokers to behave, making use of that context to automate workflows, speed up selections, and remodel operations finish‑to‑finish.

Unifying your information property with Microsoft OneLake

Each AI initiative begins with the identical basic problem: understanding the place your information lives and find out how to convey it collectively. Microsoft OneLake was constructed to resolve that drawback by unifying information throughout clouds, on-premises environments, and third-party platforms right into a single logical information lake with out pointless extracting, reworking, and loading (ETL), fragmentation, or duplicated copies.

Connecting to extra sources than ever earlier than

Right this moment, we’re increasing Mirroring in Material to assist much more methods our prospects depend on. Mirroring for SharePoint lists and Dremio at the moment are in preview with Azure Monitor coming quickly, whereas mirroring for Oracle and SAP Datasphere are usually accessible—all of which can be found as a part of the core mirroring capabilities. We’re additionally introducing prolonged capabilities in mirroring designed that will help you operationalize mirrored sources at scale, together with Change Knowledge Feed (CDF) and the flexibility to create views on prime of mirrored information, beginning with Snowflake. Prolonged capabilities for mirroring can be provided as a paid possibility.

Shortcut transformations are additionally now usually accessible, permitting information to be formed routinely because it connects to or strikes inside OneLake. You’ll be able to convert codecs reminiscent of Excel to Delta tables, now in preview, and apply AI-powered transformations.

Moreover, we’re persevering with to put money into open interoperability, guaranteeing OneLake works seamlessly with the platforms organizations already use. We’re excited to announce the flexibility to natively learn from OneLake by means of Azure Databricks Unity Catalog is now in public preview. We additionally just lately introduced the normal availability of our interoperability with Snowflake.

I’m additionally excited to share that Auger, a quickly rising provide chain platform designed to convey intelligence and automation to international operations, has constructed its platform on Material, with all information saved natively in OneLake. This structure allows Auger prospects to seamlessly entry their operations information by means of OneLake shortcuts inside their very own Material environments and use the complete energy of the platform together with Energy BI, Material information brokers, and extra. Study extra in my weblog, co-authored with Auger Chief Government Officer Dave Clark.

Defend your information with OneLake safety, now usually accessible

Safety and governance stay foundational to OneLake. I’m thrilled to announce OneLake safety can be usually accessible within the coming weeks, enabling information homeowners to outline roles, implement row- and column-level controls, and handle permissions by means of a single unified mannequin that follows the information.

To study extra about these bulletins, learn the OneLake weblog and the Material Knowledge Manufacturing unit weblog.

Processing and harmonizing information with Material analytics

AI brokers are solely as dependable as the information you feed them. Earlier than information can practice or floor an agent, it should be built-in, cleaned, and structured, so the agent operates from constant, trusted info. With industry-leading engines in Material like Spark, T-SQL, KQL, and Evaluation Providers, we are able to equip information groups to do precisely that.

Now, we’re increasing these capabilities with the introduction of Runtime 2.0 in 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 energy superior enterprise workloads. Materialized lake views are additionally now usually accessible, simplifying medallion structure implementation in Spark SQL and PySpark and enabling at all times up-to-date pipelines with no guide orchestration. As well as, a brand new agentic Copilot expertise in notebooks delivers deeper context consciousness, reasoning over your workspace, and producing code with larger velocity and precision.

For real-time eventualities, we’re launching Maps in Material into normal availability. Maps add geospatial context to your brokers and operations by turning giant volumes of location-based information into interactive, real-time visible insights.

For a complete overview of those bulletins and rather more, learn the Material Analytics weblog and the Material Actual-Time Intelligence weblog.

Creating semantic that means with Material IQ

Getting ready uncooked information for AI is crucial. The following step is reworking that information into significant, unified enterprise context. That’s the place Material IQ is available in.

Material IQ unifies analytical information and operational information, together with telemetry, time collection, graph, and geospatial information, inside a shared semantic framework of enterprise entities, relationships, properties, guidelines, and actions. As an alternative of pondering when it comes to tables and schemas, your groups and brokers can function on this framework, or ontology, aligned to how the enterprise really runs.

Material IQ ontologies will quickly change into accessible by means of an MCP server in preview, enabling brokers to find, perceive, and act on this semantic layer. Ontologies may function context sources for maps and shortly in operations brokers in Material, extending shared enterprise context straight into operational decision-making and execution.

We’re additionally excited to announce planning in Material IQ, a brand new enterprise planning functionality that allows organizations to create plans, budgets, forecasts, and state of affairs fashions straight on prime of Material’s semantic fashions. By complementing Material IQ’s ontologies with built-in planning, you get a whole, contextual view of your historic, real-time, and ahead planning information. This permits customers and brokers to shortly reply what has occurred, what is going on, and what ought to occur all from a single supply. See this in motion:

Lastly, we just lately introduced a strategic partnership with NVIDIA to energy the following era of Bodily AI by integrating Actual-Time Intelligence and Material IQ with NVIDIA Omniverse libraries. The mixed platform unifies actual‑time operational information, enterprise semantics, and bodily simulation to allow organizations to optimize their bodily operations in eventualities like clever digital twins, predictive upkeep, autonomous logistics, and vitality optimization.

To study extra about all of our companion bulletins, learn the Material ISV weblog and the planning in Material IQ weblog.

Enhancing the underlying Material IQ expertise

Powering a lot of Material IQ’s wealthy expertise is a mixture of Energy BI’s industry-leading, wealthy semantic mannequin expertise and graph in Material, our extremely scalable graph database. Already delivering insights to greater than 35 million energetic customers, semantic fashions present the best basis for coaching brokers by means of Material IQ. Now, with the final availability of Direct Lake on OneLake, your tables might be learn straight from OneLake with native safety enforcement, richer cross-item modeling, and import-class efficiency with out information motion or refresh.

I’m additionally excited to share that graph in Material can be usually accessible within the coming weeks, enabling groups to visualise and question complicated relationships throughout prospects, companions, and provide chains.

To study extra, try the Material IQ weblog and the Energy BI weblog.

Empowering brokers to behave with Material information and operations brokers

Frontier organizations are transferring past general-purpose assistants and as an alternative, adopting multi-agent methods composed of specialised brokers. These brokers are every grounded on particular information and reusable throughout completely different methods, permitting you to ship extra correct, accelerated, and scalable outcomes.

To assist your multi-agent methods, Material comes with built-in agent creation capabilities with Material information brokers and operations brokers. I’m excited to share that Material information brokers at the moment are usually accessible. Material information brokers might be regarded as digital analysts, aligned to particular area information to assist deeper evaluation and ship insights. Operations brokers complement them by monitoring real-time information, detecting patterns, and taking proactive motion.

Try a fast demo of operations brokers in Material:

These brokers can be utilized throughout Material or as foundational data sources in main AI instruments like Microsoft Foundry, Copilot Studio and even Microsoft 365 Copilot. To study extra about our AI bulletins, try the Material analytics weblog overlaying information brokers and the Material IQ weblog overlaying operations brokers.

Constructing mission-critical purposes with developer experiences in Material

Builders constructing the following era of AI purposes want a complete, cost-effective information platform that’s already built-in together with your current instruments and workflows. Right this moment, we’re increasing Material’s developer tooling to fulfill that demand.

First, Material Mannequin Context Protocol (MCP) is advancing with two main milestones. Material native MCP is now usually accessible, offering an open-source native server that connects AI coding assistants reminiscent of GitHub Copilot on to Material. Alongside this, we’re introducing the general public preview of Material distant MCP, a safe, cloud‑hosted execution engine that allows AI brokers and automation instruments to carry out authenticated actions in Material.

We’re additionally enhancing our Git integration with selective branching, permitting builders to department out for a selected characteristic and pull solely the gadgets they want. You additionally get improved change comparisons to extra simply overview current updates, and new folder relationships which present how characteristic workspaces connect with supply workspaces.

We’re additionally launching two open-source tasks to assist groups transfer sooner with Material: Agent Expertise for Material and Material Jumpstart. Agent Expertise for Material is an open-source set of purpose-built plugins that allow you to use pure language within the GitHub Copilot terminal to harness the complete energy of Microsoft Material. Moreover, Material Jumpstart is designed that will help you get off the bottom with detailed steering, reference architectures, and single‑click on deployments for pattern datasets, notebooks, pipelines, and reviews.

Lastly, we’re asserting that the Material Extensibility Toolkit (FET), an evolution of the Workload Growth Package (WDK), is now usually accessible. Together with this launch, we’re enabling assist for full CI/CD, variable library, and a brand new administration expertise within the Admin portal.

Migrating your current Azure service to Material

As Material continues to develop in performance, we’re additionally simplifying the migration from different Azure providers. Along with our current Synapse tooling, we’re bringing new migration assistants for Azure Knowledge Manufacturing unit, Azure Synapse Analytics, and Azure SQL in public preview.

The brand new Material migration assistant for Azure Knowledge Manufacturing unit and Synapse Analytics helps transfer your current pipelines and artifacts like Spark swimming pools and notebooks into Material with minimal disruption. It’s designed to assist incremental modernization, permitting groups to guage, convert, and optimize pipelines as they transition to Material. The migration assistant for SQL databases helps transfer SQL Server into Material by importing schemas by means of DACPACs, figuring out and resolving compatibility points with AI help, and guiding groups by means of evaluation and information copy workflows for a smoother cutover.

See extra Material innovation

Along with the bulletins above, we’re additionally rolling out a broad set of Material improvements throughout the platform. For a deeper have a look at the updates and what’s new this month, go to the Material March 2026 Characteristic abstract weblog, the Energy BI March 2026 characteristic abstract weblog, and the newest posts on the Material Updates channel.

Discover extra sources for Microsoft Material