Fabric Data Engineering
Run enterprise grade Spark jobs on all organizational data. Turn raw data into analytics—and AI-ready assets in OneLake.
OVERVIEW
Build and operate reliable data pipelines at scale
- Easily create a lakehouse with managed storage, folders, and Delta tables in OneLake.
- Ingest and transform large volumes of structured and unstructured data, using Fabric’s scalable Spark engine.
- Clean, join, and model data using PySpark or Spark SQL, and write ACID-compliant Delta tables.
- Run TSQL, create Power BI reports, and build AI applications—all directly on your data. SQL
- Run Spark workloads faster with the native execution engine for optimized, vectorized processing.
- Use materialized lake views to persist results and accelerate repeatable queries and pipelines.
- Automatically adjust Spark resources to workload needs and avoid paying for idle compute.
- Reduce notebook query latency with intelligent caching and optimized execution plans.
- Version notebooks and pipelines with Git to collaborate, review changes, and roll back safely.
- Promote artifacts across dev, test, and prod, and automate releases with governance.
- Develop with VS Code, notebooks, or low-code tools. Fabric adapts to your workflow.
- Work together in rich, interactive notebooks with visuals, shared resources, and built-in folders.
- Apply consistent security, access controls, and lineage across all data engineering workflows.
- Track lineage and apply governance policies across OneLake and Fabric workloads.
- Monitor Spark applications, diagnose errors, and identify bottlenecks using built-in monitoring tools.
- Create APIs over your data; connect applications and query multiple sources via GraphQL.
- Connect applications directly to governed lakehouse tables and shared datasets.
- Query lakehouse, mirrored, and external data sources through a single endpoint.
- Schedule pipelines and notebooks and integrate triggers for end-to-end automation.
- Use Copilot to generate Spark, SQL, and Python code; explain results; and accelerate notebook authoring.
- Diagnose errors, optimize queries, and get code autocomplete with performance suggestions directly in your workflow.
- Apply AI-powered functions inline in pandas or Spark to enrich data with minimal setup.
- Quickly clean and transform data using an interactive, notebook-native experience.
FEATURES
Explore demo videos
Learn how these Data Engineering capabilities help teams move faster, simplify complex data work, and unlock insights others can use.
Pay only for Spark jobs you run
Enable Autoscale Billing to run Spark jobs on serverless, pay-as-you-go compute—and pay only for job runtime and scale independently of base Fabric capacity.
CUSTOMER STORIES
Explore why organizations choose Fabric
Resources
Learn more about Fabric Data Engineering
Explore more about Fabric Data Engineering through curated blogs, hands-on tutorials, and documentation. Learn best practices, discover new capabilities, and get practical guidance to build and optimize with confidence.
Next steps
Start using Fabric for free
Try everything Fabric has to offer by signing up for a free trial—no credit card required.
Follow Microsoft Fabric