If you haven’t already, check out Arun Ulag’s hero blog “FabCon and SQLCon 2026: Unifying databases and Fabric on a single, complete platform” for a complete look at all of our FabCon and SQLCon announcements across both Fabric and our database offerings.
Coauthored by Alicia Li
Real-time starts with your operational data
Modern business applications win by reacting immediately, serving recommendations as users interact, alerting teams when anomalies occur, or updating dashboards the second business events happen. At the heart of these experiences are operational databases, where every insert, update, or delete represents a meaningful event.Historically, turning operational database changes into real-time events has been complex. Developers had to stitch together Change Data Capture (CDC) connectors, message brokers, and stream processors; understand low-level Debezium semantics and write custom code to transform raw change logs into something data analysts can easily consume. It sets higher standards for building event-driven applications, though these changes may not be advantageous for development teams.
Introducing DeltaFlow (Preview)
Fabric Eventstreams with DeltaFlow (Preview), provides a managed path from database change feeds to analytics‑ready streaming data in Microsoft Fabric. DeltaFlow natively ingests CDC events from operational databases and seamlessly transforms them into a structured, query-able shape that mirrors the source tables; so, teams can focus on building real-time applications instead of managing CDC plumbing.With DeltaFlow, Eventstreams handles raw CDC ingestion, schema registration, transformation, and destination table management for you. This removes the need to reason about Debezium payloads, variable schemas, or manual table lifecycle management, while still preserving the fidelity of database changes required for real-time analytics and automation
Transform raw Debezium events into analytics-ready form
Connect to your operational databases—Azure SQL, SQL Managed Instance (MI), SQL on VMs, and PostgreSQL—using the convenient Connector wizard. Simply select “Analytics-ready events & auto-updated schema” in the schema handling step, decide where to store the database table schemas, and that’s it!
Figure 1: Enabling DeltaFlow when connecting to an Azure SQL database.
Automatically manage destination tables as source database schema evolves
Modern applications evolve over time—new tables get added to an operational database, and existing table schemas get altered. However, these changes don’t have to interrupt your business. Where supported by the source database, DeltaFlow automatically detects changes to the source database and tables, registers the new schemas, and creates new tables on the destination. This minimizes, if not eliminates, any overheads and interruptions caused by the manual steps that are usually required to adjust schemas, manage destination tables, etc.
Figure 2: View source table schemas automatically fetched and registered in the schema registry.
Build real-time applications and dashboards using simple analytics queries
DeltaFlow produces streaming data that reflects the shape of the source tables and enriches it with essential metadata such as change type (insert, update or delete) and timestamps. This allows developers and analysts to write straightforward analytics queries—using tools like KQL—without first learning CDC‑specific semantics or parsing nested JSON payloads.
Figure 3: Raw Debezium CDC event
The GIF demonstrates the shape of the Eventhouse tables, created and managed automatically in a form reflecting the source tables. The result is a simpler path from operational data to real-time intelligence, enabling dashboards, alerts, and event-driven workflows that respond to database changes as they happen.
Figure 4: Eventhouse destination tables in the transformed, analytics-ready shape.
End-to-end example: Reacting to database changes in real time
Consider an e‑commerce application where every order update, new purchase, cancellation, or status change is written to an operational database. With DeltaFlow, these database changes are continuously captured and turned into streaming events in Fabric Eventstreams as they happen.As orders are inserted or updated, DeltaFlow transforms the underlying change feeds into analytics‑ready streams that reflect the order table schema. These streams can be routed to multiple real‑time consumers at once: a real‑time dashboard that tracks order volume and fulfillment latency, alerting logic that flags unusual spikes or failures, and downstream analytics systems that continuously aggregate metrics for operational reporting.
Because the streaming data mirrors the structure of the source tables, teams can build and evolve real-time applications using familiar analytics queries—without having to reason low-level change events or rework pipelines as schemas evolve. The result is a responsive, event-driven architecture where operational data flows seamlessly into real-time intelligence.
Get started today
DeltaFlow (Preview) is available as part of Fabric Eventstreams connectors like Azure SQL, SQL Managed Instance (MI), SQL on VM and PostgreSQL. To learn more about Fabric Eventstreams, CDC connectors and schema registry, explore the documentation.Microsoft Fabric Eventstreams Overview
Add Azure SQL Database CDC source to an eventstream
Schema Registry in Fabric Real-Time Intelligence
Real-Time Intelligence in Microsoft Fabric documentation
We’d love your feedback!
If you find this blog helpful, please give it a thumbs-up!Have ideas for what you'd like to see next? Drop us a comment or reach out to askeventstreams@microsoft.com. We’d love to hear what real-time scenarios you’re exploring and what topics you'd like us to cover in future posts.
Do you have new feature ideas? Head over to the Fabric Ideas to submit them.