Platform Architecture and Core Components: 10 practice questions

Question 1 of 10

Which description defines a data lakehouse?

1. AA governance solution providing fine-grained controls for data and AI
2. BA system combining benefits of data lakes and data warehouses
3. CA processing engine whose compute resources also provide the storage layer
4. DAn optimized storage layer limited to enforcing table schemas

Question 2 of 10

Which core component supplies ACID transactions and schema enforcement?

1. AThe raw ingestion layer before files become tables
2. BApache Spark as the scalable processing engine
3. CUnity Catalog as the unified governance layer
4. DDelta Lake as the optimized storage layer

Question 3 of 10

Which core component provides unified, fine-grained governance for data and AI?

1. AStructured Streaming
2. BDelta Lake
3. CApache Spark
4. DUnity Catalog

Question 4 of 10

How does Apache Spark fit the Databricks lakehouse architecture?

1. AIt runs as the governance engine on compute coupled permanently to storage
2. BIt acts as the optimized storage layer that supplies ACID transactions
3. CIt runs as the scalable processing engine on compute decoupled from storage
4. DIt registers tables and applies access control lists for governed data

Question 5 of 10

What can check for missing or unexpected data when files become Delta tables?

1. ADelta Lake schema enforcement
2. BUnity Catalog access control lists
3. CApache Spark compute separation
4. DUnity Catalog lineage tracking

Question 6 of 10

A team needs to follow data as it is transformed and refined. Which capability addresses this need?

1. AUnity Catalog lineage tracking
2. BSpark processing with separated storage
3. CDelta Lake schema enforcement
4. DDelta Lake ACID transaction support

Question 7 of 10

Which relationship helps provide the foundations for Lakeflow pipelines and Auto Loader?

1. ADelta Lake replaces the need for a processing engine
2. BUnity Catalog replaces Structured Streaming during ingestion
3. CApache Spark couples compute permanently to Delta storage
4. DStructured Streaming integrates tightly with Delta Lake

Question 8 of 10

A transformation workload needs massively scalable processing, transactional table reliability, and compute that remains independently scalable from storage. Which arrangement meets all three constraints?

1. AUse Structured Streaming with Unity Catalog and couple compute permanently to storage
2. BUse Delta Lake with Unity Catalog and omit a scalable processing engine
3. CUse Spark on decoupled compute with Unity Catalog as the transactional storage layer
4. DUse Spark on decoupled compute with Delta Lake as the transactional storage layer

Question 9 of 10

A pipeline receives batch and streaming files, must detect unexpected fields after table conversion, and must preserve lineage through refinement. Which design meets both control requirements?

1. AUse Unity Catalog lineage and Spark transformations while leaving files outside Delta tables
2. BUse Delta transactions and Unity Catalog privileges as a substitute for transformation lineage
3. CUse Delta schema checks and Spark transformations without Unity Catalog lineage
4. DUse Delta schema enforcement and register the tables in Unity Catalog for lineage

Question 10 of 10

An organization wants one source of truth for engineering, machine learning, and reporting, with traceability from served data back through refinement. Which arrangement best fits?

1. AOne lakehouse with optimized layouts but separate governance for each workload
2. BOne lakehouse with optimized layouts and a unified governance model for lineage
3. COne Delta storage layer with Spark processing but no lineage governance
4. DSeparate lakehouses for each workload with a shared Unity Catalog governance model