You work for MDFT Pro, a well-known training agency that tracks student learning activity in real-time. Mark, a Data Platform Architect at MDFT Pro, manages a Fabric workspace containing a large table called StudentActivity with 2 billion rows storing historical student engagement data. The company’s learning management system generates a change file every 30 minutes containing only the activities that occurred since the last file was generated (new course views, completed lessons, forum posts, etc.). Mark is designing a data pipeline that will automatically process each change file when it’s generated and load the contents into the StudentActivity table. The solution must efficiently handle two types of operations: creating new records for student activities and deleting records when students withdraw from courses. Additionally, the system must support versioning to maintain a complete audit trail of all changes over time, allowing analysts to query historical states of the data. Mark needs to select the appropriate loading pattern for each operation type.
What loading pattern should Mark use for deleting existing records?
Choose the correct answer from the options below.
Explanations for each answer: