An Azure service for ingesting, preparing, and transforming data at scale.
Hello !
Thank you for posting on Microsoft Learn.
You need to install the on-prem data gateway (Enterprise) on a server that can reach your SQL Server. Sign in with your Fabric/Power BI tenant account and make sure that the outbound TCP 443 is open to Microsoft endpoints.
In Fabric, follow this path : Manage → Connections → New → SQL Server → select your Gateway → enter server & DB → choose auth (SQL Auth or Windows Entra ID via Kerberos SSO if configured) → Test.
In your data pipeline, in your copy data activity:
- source: your SQL Server connection (table or custom query).
- sink: Lakehouse (files or tables/delta) or warehouse. You can enable “auto-create table” if loading to Lakehouse tables.
You can enable partitioned copy on an integer/date key, set batch size, and map types and schedule the pipeline and monitor runs in pipeline monitoring.
The gateway is the Fabric equivalent of ADF SHIR for on-prem. Keep the gateway host always on throughput/concurrency is limited by that host.