100% Latest & Updated Databricks Certified Data Engineer Associate Practice Test Questions, Exam Dumps & Verified Answers!
30 Days Free Updates, Instant Download!
Question 30
An SQL transformation has sporadic traffic and the team wants reduced infrastructure administration. Which action should the engineer perform to establish the intended control?
- Record the correct Delta version and run a tested RESTORE to that version.
- Retain immutable raw events in bronze and write cleaned records to a new silver table.
- Select serverless warehouse settings and validate workspace eligibility and permissions.
- Create catalog and schema objects and register the datasets under them.
Explanation
Correct Answer
C:Select serverless warehouse settings and validate workspace eligibility and permissions. The query starts without engineers managing the underlying cluster nodes.
Incorrect Answers
A:An analyst needs to revert a faulty data batch without restoring a whole storage snapshot. In those circumstances, “Record the correct Delta version and run a tested RESTORE to that version” fits. Select serverless warehouse settings and validate workspace eligibility and permissions.
B:When a team wants to keep raw source records while refining validated data for analytics, “Retain immutable raw events in bronze and write cleaned records to a new silver table” could be relevant. Select serverless warehouse settings and validate workspace eligibility and permissions.
D:A business unit wants tables discoverable and consistently governed by separate engineering workspaces. In those circumstances, “Create catalog and schema objects and register the datasets under them” fits. Select serverless warehouse settings and validate workspace eligibility and permissions.
Question 31
An SQL transformation has sporadic traffic and the team wants reduced infrastructure administration. Which observation best proves the workload meets its requirement?
- The query starts without engineers managing the underlying cluster nodes.
- Concurrent writers commit valid Delta transactions and readers see consistent snapshots.
- The table's rows match the selected earlier version after restoration.
- Users with appropriate grants can find the same three-level table name from authorized workspaces.
Explanation
Correct Answer
A:The query starts without engineers managing the underlying cluster nodes.
Incorrect Answers
B:If a compliance team wants transactional updates to shared analytical files without handing each application a custom locking protocol, “Concurrent writers commit valid Delta transactions and readers see consistent snapshots” may be the appropriate response. The query starts without engineers managing the underlying cluster nodes.
C:An analyst needs to revert a faulty data batch without restoring a whole storage snapshot. In those circumstances, “The table's rows match the selected earlier version after restoration” fits. The query starts without engineers managing the underlying cluster nodes.
D:When a business unit wants tables discoverable and consistently governed by separate engineering workspaces, “Users with appropriate grants can find the same three-level table name from authorized workspaces” could be relevant. The query starts without engineers managing the underlying cluster nodes.
Question 32
An SQL transformation has sporadic traffic and the team wants reduced infrastructure administration. The team observes: The team cannot use the proposed warehouse mode despite a valid SQL statement. Which underlying cause most directly explains this failure?
- The pipeline replaced cloud objects outside Delta's transaction protocol.
- The workspace or selected workload is ineligible for the chosen serverless capability.
- Relevant Delta history was no longer retained under the table's retention policy.
- Writers bypassed Delta transaction management and modified data files directly.
Explanation
Correct Answer
B:The workspace or selected workload is ineligible for the chosen serverless capability. Select serverless warehouse settings and validate workspace eligibility and permissions.
Incorrect Answers
A:If two applications must query a lakehouse table during an ETL merge without being exposed to partial updates, “The pipeline replaced cloud objects outside Delta's transaction protocol” may be the appropriate response. The workspace or selected workload is ineligible for the chosen serverless capability.
C:When an analyst needs to revert a faulty data batch without restoring a whole storage snapshot, “Relevant Delta history was no longer retained under the table's retention policy” could be relevant. The workspace or selected workload is ineligible for the chosen serverless capability.
D:If a compliance team wants transactional updates to shared analytical files without handing each application a custom locking protocol, “Writers bypassed Delta transaction management and modified data files directly” may be the appropriate response. The workspace or selected workload is ineligible for the chosen serverless capability.
Question 33
A data scientist runs an interactive notebook with iterative Spark transformations rather than pure dashboard SQL. Which design choice best addresses the stated business and technical requirement?
- Use Delta table version history and supported time travel or RESTORE.
- Separate bronze ingestion, silver conformance, and gold serving layers.
- Use Unity Catalog as the common metastore and authorization plane.
- Select compatible notebook compute for interactive Spark execution.
Explanation
Correct Answer
D:Select compatible notebook compute for interactive Spark execution. PySpark cells run in the notebook and can read the authorized Unity Catalog tables.
Incorrect Answers
A:An analyst needs to revert a faulty data batch without restoring a whole storage snapshot. In those circumstances, “Use Delta table version history and supported time travel or RESTORE” fits. Select compatible notebook compute for interactive Spark execution.
B:When a team wants to keep raw source records while refining validated data for analytics, “Separate bronze ingestion, silver conformance, and gold serving layers” could be relevant. Select compatible notebook compute for interactive Spark execution.
C:If a business unit wants tables discoverable and consistently governed by separate engineering workspaces, “Use Unity Catalog as the common metastore and authorization plane” may be the appropriate response. Select compatible notebook compute for interactive Spark execution.
Question 34
A data scientist runs an interactive notebook with iterative Spark transformations rather than pure dashboard SQL. Which setup step most directly realizes the proposed design?
- Perform the update with a Delta MERGE transaction instead of overwriting underlying files.
- Write the dataset as Delta and register the table in Unity Catalog.
- Attach the notebook to supported interactive compute with the needed runtime and access mode.
- Retain immutable raw events in bronze and write cleaned records to a new silver table.
Explanation
Correct Answer
C:Attach the notebook to supported interactive compute with the needed runtime and access mode. PySpark cells run in the notebook and can read the authorized Unity Catalog tables.
Incorrect Answers
A:When two applications must query a lakehouse table during an ETL merge without being exposed to partial updates, “Perform the update with a Delta MERGE transaction instead of overwriting underlying files” could be relevant. Attach the notebook to supported interactive compute with the needed runtime and access mode.
B:If a compliance team wants transactional updates to shared analytical files without handing each application a custom locking protocol, “Write the dataset as Delta and register the table in Unity Catalog” may be the appropriate response. Attach the notebook to supported interactive compute with the needed runtime and access mode.
D:When a team wants to keep raw source records while refining validated data for analytics, “Retain immutable raw events in bronze and write cleaned records to a new silver table” could be relevant. Attach the notebook to supported interactive compute with the needed runtime and access mode.
Question 35
A data scientist runs an interactive notebook with iterative Spark transformations rather than pure dashboard SQL. What is the most relevant verification evidence for this case?
- A concurrent SELECT observes one committed snapshot rather than mixed intermediate records.
- A malformed bronze record remains traceable while gold aggregates exclude it under quality rules.
- Concurrent writers commit valid Delta transactions and readers see consistent snapshots.
- PySpark cells run in the notebook and can read the authorized Unity Catalog tables.
Explanation
Correct Answer
D:PySpark cells run in the notebook and can read the authorized Unity Catalog tables.
Incorrect Answers
A:If two applications must query a lakehouse table during an ETL merge without being exposed to partial updates, “A concurrent SELECT observes one committed snapshot rather than mixed intermediate records” may be the appropriate response. PySpark cells run in the notebook and can read the authorized Unity Catalog tables.
B:A team wants to keep raw source records while refining validated data for analytics. In those circumstances, “A malformed bronze record remains traceable while gold aggregates exclude it under quality rules” fits. PySpark cells run in the notebook and can read the authorized Unity Catalog tables.
C:When a compliance team wants transactional updates to shared analytical files without handing each application a custom locking protocol, “Concurrent writers commit valid Delta transactions and readers see consistent snapshots” could be relevant. PySpark cells run in the notebook and can read the authorized Unity Catalog tables.
Question 36
A data scientist runs an interactive notebook with iterative Spark transformations rather than pure dashboard SQL. The team observes: PySpark cells fail even though the SQL dashboard warehouse answers SQL queries. What is the most likely mechanism behind the observed defect?
- The pipeline replaced cloud objects outside Delta's transaction protocol.
- The notebook was not attached to an execution environment supporting its Spark workload.
- Writers bypassed Delta transaction management and modified data files directly.
- The table was stored only in a workspace-local legacy metastore rather than the shared Unity Catalog.
Explanation
Correct Answer
B:The notebook was not attached to an execution environment supporting its Spark workload. Attach the notebook to supported interactive compute with the needed runtime and access mode.
Incorrect Answers
A:Two applications must query a lakehouse table during an ETL merge without being exposed to partial updates. In those circumstances, “The pipeline replaced cloud objects outside Delta's transaction protocol” fits. The notebook was not attached to an execution environment supporting its Spark workload.
C:If a compliance team wants transactional updates to shared analytical files without handing each application a custom locking protocol, “Writers bypassed Delta transaction management and modified data files directly” may be the appropriate response. The notebook was not attached to an execution environment supporting its Spark workload.
D:A business unit wants tables discoverable and consistently governed by separate engineering workspaces. In those circumstances, “The table was stored only in a workspace-local legacy metastore rather than the shared Unity Catalog” fits. The notebook was not attached to an execution environment supporting its Spark workload.
Question 37
A point-of-sale system publishes a new immutable object-store file every five minutes. Which approach most directly satisfies the requirement?
- Use a bounded REST extraction in a scheduled notebook with explicit pagination and retry handling.
- Use incremental file ingestion instead of rescanning an entire historical directory each run.
- Use COPY INTO against the Delta target for incremental file loading from storage.
- Use a notebook-based JDBC extraction through a governed ingestion job.
Explanation
Correct Answer
B:Use incremental file ingestion instead of rescanning an entire historical directory each run. A newly arriving object is processed while previously checkpointed objects are not redundantly ingested.
Incorrect Answers
A:When an external service exposes a paginated REST endpoint with rate limits and no managed connector, “Use a bounded REST extraction in a scheduled notebook with explicit pagination and retry handling” could be relevant. Use incremental file ingestion instead of rescanning an entire historical directory each run.
C:An operations team wants a SQL-only one-time load of historical CSV objects stored in an approved S3 prefix. In those circumstances, “Use COPY INTO against the Delta target for incremental file loading from storage” fits. Use incremental file ingestion instead of rescanning an entire historical directory each run.
D:When a niche on-premises database lacks a supported managed connector but allows JDBC reads, “Use a notebook-based JDBC extraction through a governed ingestion job” could be relevant. Use incremental file ingestion instead of rescanning an entire historical directory each run.
Question 38
A point-of-sale system publishes a new immutable object-store file every five minutes. Which configuration or engineering action would most directly implement the requirement?
- Restore the original checkpointLocation rather than creating an empty path.
- Configure an Auto Loader stream with a persistent checkpoint for this directory.
- Use Auto Loader discovery and a durable progress checkpoint for incremental processing.
- Keep the COPY INTO target and source identity stable across reruns.
Explanation
Correct Answer
B:Configure an Auto Loader stream with a persistent checkpoint for this directory. A newly arriving object is processed while previously checkpointed objects are not redundantly ingested.
Incorrect Answers
A:If a streaming ingestion restarts after maintenance and must continue from its last committed progress, “Restore the original checkpointLocation rather than creating an empty path” may be the appropriate response. Configure an Auto Loader stream with a persistent checkpoint for this directory.
C:When a large object-store drop includes millions of files with continual arrivals, “Use Auto Loader discovery and a durable progress checkpoint for incremental processing” could be relevant. Configure an Auto Loader stream with a persistent checkpoint for this directory.
D:If a data owner receives daily Parquet exports and wants previously copied files skipped on reruns, “Keep the COPY INTO target and source identity stable across reruns” may be the appropriate response. Configure an Auto Loader stream with a persistent checkpoint for this directory.
Question 39
A point-of-sale system publishes a new immutable object-store file every five minutes. Which test would establish that the change worked as intended?
- The ingested tables are discoverable under the approved UC namespace and grants.
- The stream stops for an unapproved added column until the schema is updated.
- Previously unseen optional keys do not erase the original raw record.
- A newly arriving object is processed while previously checkpointed objects are not redundantly ingested.
Explanation
Correct Answer
D:A newly arriving object is processed while previously checkpointed objects are not redundantly ingested.
Incorrect Answers
A:An audit team needs stable ownership of data ingested from a managed enterprise application. In those circumstances, “The ingested tables are discoverable under the approved UC namespace and grants” fits. A newly arriving object is processed while previously checkpointed objects are not redundantly ingested.
B:When a customer source cannot accept newly introduced columns without explicit schema approval, “The stream stops for an unapproved added column until the schema is updated” could be relevant. A newly arriving object is processed while previously checkpointed objects are not redundantly ingested.
C:If a telemetry source emits JSON lines with heterogeneous optional keys, “Previously unseen optional keys do not erase the original raw record” may be the appropriate response. A newly arriving object is processed while previously checkpointed objects are not redundantly ingested.
Question 40
A point-of-sale system publishes a new immutable object-store file every five minutes. The team observes: Restarting a pipeline causes historical objects to be treated as new data. Which configuration or design gap best explains the symptom?
- The job uses Auto Loader cloudFiles against a broker-only event source.
- The stream's checkpoint was replaced rather than reused for the same source.
- A handwritten REST poller was used instead of the supported managed connector.
- A continuously active stream was chosen for an essentially one-off load.
Explanation
Correct Answer
B:The stream's checkpoint was replaced rather than reused for the same source. Configure an Auto Loader stream with a persistent checkpoint for this directory.
Incorrect Answers
A:When a real-time event feed has no cloud-object landing area and is delivered over an event broker, “The job uses Auto Loader cloudFiles against a broker-only event source” could be relevant. The stream's checkpoint was replaced rather than reused for the same source.
C:A CRM administrator wants managed incremental ingestion into governed Databricks tables without writing custom REST pagination. In those circumstances, “A handwritten REST poller was used instead of the supported managed connector” fits. The stream's checkpoint was replaced rather than reused for the same source.
D:When a static set of archive files must be loaded once through SQL with minimal pipeline machinery, “A continuously active stream was chosen for an essentially one-off load” could be relevant. The stream's checkpoint was replaced rather than reused for the same source.
Certified Data Engineer Associate Practice Tests