What is the primary operational reason for pausing a Presto engine in watsonx.data when it is not in use?

Pausing an engine in watsonx.data stops compute billing for that engine; since storage and compute are decoupled; data is safe in object storage and queries resume immediately when the engine is started again. The incorrect options ("To apply schema changes to tables managed by the engine", "To release metastore locks held by long-running queries", "To flush the Presto query cache and improve performance for the next session") are distractors that don't fully capture the concept described.