Note the following guidelines and considerations for using Microsoft Fabric Lakehouse targets:
•Microsoft Fabric Lakehouse targets connect through a Spark driver. Initializing a Spark driver carries significant overhead, so connection setup time can affect overall job performance, particularly under load. When running jobs that write to a Microsoft Fabric Lakehouse target, monitor latency and connection overhead across all job modes. Avoid designs that open and close connections frequently and consolidate writes into fewer, larger operations to spread the connection cost across more data.
•Database ingestion and replication jobs that have Microsoft Fabric Lakehouse targets do not allow schema drift with the Modify Column, Drop Colum, and Rename Column options.
•The Microsoft Fabric Lakehouse target does not allow capital letters in table names. All tables are created with lowercase letters by default.