Moving data into Fabric with a data pipeline

Fabric Data Factory pipelines let you orchestrate and automate data ingestion from a wide range of on-premises and cloud sources into Fabric using Copy Data activities and other pipeline activities. Pipelines are built from activities (Copy, Dataflow, Notebook, Stored Procedure, etc.) connected in a directed workflow with parameters, triggers, and control-flow logic like ForEach, If Condition, and Until. They provide a low-code, visual authoring experience similar to Azure Data Factory/Synapse pipelines, since Fabric pipelines share the same underlying engine.

1 · Learn the must-know

• 1The Copy activity is the primary tool for ingesting data at scale into Fabric destinations such as Lakehouse tables/files, Warehouse tables, and KQL databases, and supports schema mapping, partitioning, and staging for performance.
• 2Connections to sources use connectors and linked-service-like connection objects; on-premises or private-network sources require an On-premises Data Gateway or a Virtual Network (VNet) data gateway.
• 3Pipelines can be triggered on a schedule, via tumbling window, event-based triggers (e.g., storage events), or on-demand/manual execution, and can be invoked from another pipeline using the Invoke Pipeline activity.
• 4Parameters and variables allow pipelines to be reused across environments/sources, while pipeline expressions (dynamic content) let you reference activity outputs and system variables at runtime.
• 5Copy Data assistant provides a wizard-driven experience to quickly configure source, destination, and mapping for common ingestion scenarios without manually building the pipeline canvas.
• 6Monitoring is available through the Fabric Monitoring hub and pipeline run history, which shows activity-level status, duration, and error details for troubleshooting failed or long-running ingestions.

2 · Check your understanding