Global view vs Temp view in the context of Databricks

Temp View

• Available within the context of a single notebook.
• Can be used to share data across different language REPLs within the same notebook (e.g., Python to Scala).
• Limited to the current Spark session.
• Not prefixed with a namespace; accessed directly by the view name.

Global Temp View

• Available to all notebooks running on the same Databricks cluster.
• Shared across multiple Spark sessions within the same Spark application.
• Prefixed with the global_temp namespace; accessed as global_temp.view_name.
• Lost when the Spark application (or Databricks cluster) stops.