Skip to content

Fabric foundations

Microsoft Fabric is a SaaS analytics platform that combines familiar Power BI, Azure Synapse Analytics, and Azure Data Factory experiences. It gives teams a shared data estate through OneLake while retaining specialized experiences for engineering, warehousing, science, real-time analytics, and reporting.

Workloads in the workshop

Workload Workshop relevance
Data Factory Ingest and orchestrate data movement and transformation.
Data Engineering Use Spark-based notebooks to prepare and transform data.
Data Warehouse Query structured, curated data with SQL analytics.
Data Science Train, evaluate, and register models with MLflow patterns.
Power BI Build reports and semantic models from governed data.
Data Activator React to data events with operational automation.

OneLake and storage choices

OneLake reduces unnecessary data copies by providing a shared logical lake for the organization. In the workshop, use the lakehouse flow for raw-to-curated transformations; select warehouses where relational SQL reporting is the natural interface.

flowchart TB O[Operational and external sources] --> F[Fabric ingestion] F --> L[OneLake] L --> LH[Lakehouse and Spark] L --> WH[Warehouse and SQL] LH --> BI[Power BI] WH --> BI

Concepts to recognize

  • Parquet and Delta: open table formats used for efficient analytical storage and reliable table operations.
  • Shortcuts and mirroring: connect to data without treating every scenario as a full copy-and-paste migration.
  • V-Order and Z-Order: storage and data-layout optimizations that can improve analytical performance for suitable workloads.
  • Purview integration: governance, discovery, security, lineage, and compliance remain part of the data platform lifecycle.

See the source Fabric overview and Microsoft Fabric documentation for current product details.