The Microsoft Fabric Blog team highlights recent innovations in Fabric Data Factory, focusing on enterprise-scale data integration, AI-driven features, and new performance updates to empower data engineers and analysts.

Unifying Data Estates with Microsoft Fabric Data Factory: AI, Integration, and Innovation

Overview

Microsoft Fabric Data Factory now brings together industry-leading, cloud-first data integration capabilities built on OneLake. By fusing Azure Data Factory’s pro-grade integration with Power Query’s citizen data transformation features, Fabric Data Factory offers a unified approach to connect, prepare, and manage data from multicloud and on-premises sources.

Key Feature Updates

Dataflow Gen2: Pricing and Performance Improvements

  • Lower base rate (12 CU, down from 16 CU)
  • Tiered pricing for long-running jobs (reduced cost after 10 minutes)
  • Improved performance via modern PQ evaluator and partitioned query execution
  • New design-time experience with fast ‘Preview only Steps’ for editing queries

Petabyte-Scale Data Movement

  • 170+ connectors for diverse data sources/destinations
  • Copy job allows high-throughput, cross-cloud movement, including AWS, Oracle, PostgreSQL, and more
  • Latest features: Copy job orchestration in pipelines, variable library for environment parameterization, support for new formats like Iceberg/JSON, improved scheduling, and integration with Lakehouse, SharePoint, Snowflake
  • Enhanced incremental data ingestion (Change Data Feeds, CDC)

Pro-Code & Low-Code Orchestration

  • Low-code pipelines for robust scheduling and automation
  • CI/CD enhancements: ADF pipeline migration utility, variable libraries, monitoring workspace
  • Pro-code integration with managed Airflow for serverless DAG orchestration
  • New activities: Email, Teams, SPN/Workspace identity, Dataflow parameterization

Database Mirroring

  • Zero-copy, zero-ETL data replication into OneLake from major sources (Google BigQuery, Oracle, Azure SQL, Snowflake, etc.)
  • Secure access via Private Link, on-prem gateway support, and semantic modeling direct from mirrored databases

AI-Powered Data Integration (Copilot)

  • Author, debug, and monitor pipelines using AI copilots
  • New features: Natural language to define columns and transformations, automate documentation, and chat-driven pipeline creation
  • Upcoming: AI-powered transformation prompts for sentiment analysis, summarization, and categorization

Security and Mission-Critical Enhancements

  • Workspace identity support and Private Link for secure, credential-free connections
  • Azure Key Vault integration for managing secrets
  • VNet Gateway, API support, and enhanced administration via PowerShell
  • Planned: Snowflake key-pair authentication for automation

Practical Impact

  • Simplifies complex data estate unification, enabling data engineers and analysts to build intelligence-driven workflows
  • Reduces operational costs while increasing performance and scalability
  • Empowers users of all skill levels (pro-code, low-code) to automate and transform data across clouds
  • Accelerates AI and analytics adoption with integrated copilots and efficient dataflows

Resources

This update underlines Microsoft’s ongoing commitment to community-driven innovation in the Fabric ecosystem. For more in-depth feature guides, the Fabric Blog will publish detailed follow-ups.

This post appeared first on “Microsoft Fabric Blog”. Read the entire article here