LFC Cheat Sheet 2026

The 30 highest-yield LFC facts, distilled from real exam questions. Print it, save it as a PDF, or study it here — free, no sign-up.

1. When using Unity Catalog, what is the correct three-level namespace format to fully qualify a table? → catalog.schema.table
2. What is the default behavior of Databricks SQL when a query references a table that does not exist in the current catalog? → It raises a TABLE_OR_VIEW_NOT_FOUND error
3. Which of the following best describes a key advantage of keeping the Bronze layer immutable? → It enables full reprocessing and auditing from the original source data
4. What is the primary ethical obligation of a LFC professional when a conflict of interest arises during photon engine optimization activities? → Disclose the conflict to all relevant parties and recuse from the decision if necessary
5. A data steward applies a column mask to a PII column in Unity Catalog. What happens when a user without the 'unmask' privilege queries that column? → The masking function returns a substituted or redacted value
6. Which Spark join strategy avoids a shuffle by broadcasting the smaller table to all executor nodes? → Broadcast hash join
7. Which Unity Catalog privilege allows a user to read data from a table but explicitly prevents them from viewing the table's column definitions? → SELECT
8. In a Lakehouse, compute and storage are typically described as being decoupled. What is the primary operational benefit of this design? → Compute resources can be scaled independently and shut down without losing data
9. A LFC professional encounters an unfamiliar situation while performing apache spark data engineering duties. What is the most appropriate first action? → Consult relevant standards, guidelines, or a qualified supervisor before proceeding
10. Which Medallion layer typically applies data quality rules such as rejecting records with null primary keys? → Silver
11. When using Unity Catalog, what happens if a user runs SELECT * on a table where a column mask is applied and they do not meet the mask's condition? → The masked column returns NULL or a transformed value instead of the real data
12. What is the role of data versioning in lakehouse systems? → To track and revert changes
13. What does setting `registered_model_name` in `mlflow.log_model()` accomplish? → It logs and simultaneously registers the model in the registry in one step
14. What is the significance of the 'photonized' metric shown in Databricks cluster metrics for a running query? → It indicates the fraction of CPU time spent in Photon-native execution vs. JVM Spark
15. In Databricks Unity Catalog, which permission is required to register a new model in a schema? → CREATE MODEL
16. When processing a Kafka stream in Structured Streaming, the raw message value is read as binary. Which function is typically used to convert it to a string? → `cast(value as string)` or `col('value').cast('string')`
17. When ingesting JSON data with a schema that may evolve, which Spark option should you enable to handle new fields without failing? → mergeSchema
18. Which SQL construct in Databricks allows you to iterate over a result set row by row inside a stored procedure or scripting block? → FOR loop with a query expression
19. Which Delta Lake table feature works synergistically with Photon to minimize data scanned during queries? → Column-level statistics stored in the Delta transaction log
20. Which Spark operation should you use to merge new CDC (change data capture) records into an existing Delta table using an upsert pattern? → MERGE INTO with MATCHED and NOT MATCHED clauses
21. In Unity Catalog's three-level namespace, what is the correct order of hierarchy? → Catalog → Schema → Table
22. Which DLT pipeline execution mode processes all available data and then stops? → Triggered mode
23. What is the effect of calling `client.delete_registered_model(name)` in MLflow? → Deletes the registered model and ALL its versions from the registry
24. Which ACID property ensures that a Delta Lake table never contains partial results from a failed MERGE operation? → Atomicity
25. What is the recommended approach when a source system changes its schema and sends new fields to the Bronze layer? → Allow schema evolution so Bronze absorbs the new fields without pipeline failure
26. In Databricks SQL, which privilege must be granted on a SQL Warehouse to allow a user to execute queries against it? → CAN USE
27. What is the effect of setting the 'spot_policy' to 'COST_OPTIMIZED' on a Databricks SQL Warehouse? → The warehouse preferentially uses spot/preemptible instances to reduce compute cost
28. What is the recommended approach for managing data access for a team of 50 analysts in Unity Catalog? → Create a group, add users to it, and grant privileges to the group
29. When you call `spark.read.format('delta').load('/path')`, what does Spark consult first to determine which files to read? → The Delta transaction log (_delta_log)
30. What does 'schema-on-read' mean in a lakehouse ETL context? → The schema is applied and interpreted when data is queried, not at ingestion