View Full Microsoft DP-750 Exam Dumps and Practice Test Dumps.

 

Question 381

A data engineer needs to create a logical namespace inside a Unity Catalog catalog for organizing related tables and views. Which object should be created?

1. Volume
2. Schema
3. Storage credential
4. External location

Correct Answer: 2

Explanation

A schema provides a logical namespace within a Unity Catalog catalog and is used to organize objects such as tables, views, and other supported data assets. Organizations can create separate schemas for different business domains, applications, environments, or teams. This hierarchical organization makes data discovery and permission management easier. A volume is intended for governed file storage, a storage credential provides authentication for external storage, and an external location combines a storage path with a credential for governed access. Therefore, when the requirement is to create an organizational namespace inside a catalog for related data objects, a schema is the appropriate Unity Catalog object.

Question 382

A pipeline must process a large volume of incoming files continuously as they arrive in cloud storage. Which Databricks capability is specifically designed for scalable incremental file ingestion?

1. Auto Loader
2. DESCRIBE HISTORY
3. SQL warehouse
4. Column masking

Correct Answer: 1

Explanation

Auto Loader is designed for incremental and scalable ingestion of files arriving in cloud object storage. Instead of repeatedly processing the entire directory, Auto Loader can identify newly arriving files and process them as part of an ingestion pipeline. It also provides capabilities for schema inference and schema evolution, which are useful when source structures change over time. DESCRIBE HISTORY is used to inspect Delta table transactions, a SQL warehouse provides SQL compute, and column masking protects sensitive values. Therefore, when a large and continuously growing collection of cloud-storage files needs to be ingested incrementally, Auto Loader is the appropriate Databricks capability.

Question 383

A data engineer wants to make sure that a downstream task does not execute until an upstream transformation has completed successfully. Which workflow feature should be configured?

1. Data skipping
2. Job dependency
3. Column masking
4. Table retention

Correct Answer: 2

Explanation

A job dependency establishes an execution relationship between workflow tasks. When a downstream task depends on an upstream task, the workflow can wait for the required upstream task to complete successfully before allowing the dependent task to execute. This is important for multi-step pipelines where later transformations require outputs generated by earlier processing stages. Data skipping improves query efficiency, column masking protects sensitive information, and table retention controls how long historical data or files are retained. These features do not determine workflow execution order. Therefore, when one task must wait for another task to finish successfully, a job dependency should be configured.

Question 384

A Delta table is queried frequently using filters on specific columns. The engineer wants to organize the table data to improve data skipping for these access patterns. Which capability should be considered?

1. Secret management
2. User provisioning
3. Liquid clustering
4. Job retries

Correct Answer: 3

Explanation

Liquid clustering can organize Delta table data according to selected clustering keys and is designed to adapt as data and query patterns change. When queries frequently filter on appropriate columns, clustering can improve data organization and help the engine avoid unnecessary file reads through more effective data skipping. This can be useful for large datasets where traditional physical partitioning may not be ideal. Secret management handles sensitive credentials, user provisioning manages identities, and job retries control recovery from failed workflow tasks. Therefore, when a Delta table needs an adaptive physical layout optimized around frequently queried columns, liquid clustering is a relevant capability.

Question 385

A data engineer needs to apply a custom transformation to every micro-batch before writing the results to a target Delta table. Which Structured Streaming feature should be used?

1. VACUUM
2. foreachBatch
3. SHOW GRANTS
4. DESCRIBE DETAIL

Correct Answer: 2

Explanation

foreachBatch allows custom processing logic to be applied to each micro-batch generated by a Structured Streaming query. This is useful when a streaming pipeline needs to perform operations that are easier to express using batch-oriented DataFrame logic. For example, a pipeline can use foreachBatch to process each micro-batch and perform customized writes or synchronization operations against a Delta target. VACUUM manages obsolete files, SHOW GRANTS displays permissions, and DESCRIBE DETAIL provides table metadata. These commands do not provide per-micro-batch custom processing. Therefore, foreachBatch is the appropriate feature when custom logic must be applied to every streaming micro-batch.

Question 386

A company wants to prevent unauthorized users from seeing rows belonging to another business region. Which Unity Catalog governance capability is designed for this requirement?

1. Row filtering
2. Autoscaling
3. File compaction
4. Git version control

Correct Answer: 1

Explanation

Row filtering can restrict the records returned to a user based on defined rules. This supports row-level access-control scenarios where different users or groups should see different subsets of the same table. For example, a regional user can be restricted to records associated with an authorized region. Autoscaling manages compute capacity, file compaction improves storage layout, and Git version control tracks source-code changes. These capabilities do not determine which table rows a user can access. Therefore, when users must be prevented from seeing records outside their authorized business region, row filtering is the appropriate Unity Catalog governance capability.

Question 387

A data engineer wants to investigate which operations changed a Delta table during the previous day. Which command should be used?

1. CREATE SCHEMA
2. DESCRIBE HISTORY
3. OPTIMIZE
4. SHOW VOLUMES

Correct Answer: 2

Explanation

DESCRIBE HISTORY provides the transaction history of a Delta table and can be used to investigate operations performed against it. The history can contain information about operations such as writes, updates, deletes, merges, and other table changes. This makes it useful for auditing, troubleshooting, and understanding when changes occurred. CREATE SCHEMA creates a namespace, OPTIMIZE improves Delta file organization, and SHOW VOLUMES relates to volume discovery. These commands do not provide the same transaction-level history of a Delta table. Therefore, when an engineer needs to investigate previous table operations, DESCRIBE HISTORY is the appropriate command.

Question 388

A streaming application must tolerate records that arrive several minutes after their expected event time. Which configuration is most relevant?

1. Cluster policy
2. Watermark
3. External location
4. Storage credential

Correct Answer: 2

Explanation

A watermark defines how much event-time lateness a Structured Streaming application is prepared to handle for supported stateful operations. If events arrive several minutes late, the watermark can provide a defined period during which those late events may still be considered. As event-time progress advances, older state can eventually be removed, helping control resource consumption. Cluster policies govern compute configurations, external locations govern cloud storage paths, and storage credentials provide authentication for those paths. These features do not manage late-arriving streaming records. Therefore, when a streaming application needs to tolerate a defined amount of event-time delay, watermarking is the relevant configuration.

Question 389

A data engineer needs to ensure that a table rejects records when a required numeric value falls outside an allowed range. Which table-level mechanism can enforce this rule?

1. Delta constraint
2. SQL warehouse
3. Git branch
4. External location

Correct Answer: 1

Explanation

A Delta constraint can enforce data-quality rules directly at the table level. A check constraint can specify conditions that incoming records must satisfy, such as requiring a numeric value to remain within an allowed range. If an attempted write violates the applicable constraint, the operation can fail rather than allowing invalid data into the table. A SQL warehouse provides SQL compute, a Git branch manages source-code changes, and an external location governs access to cloud storage. These features do not enforce row-level data-quality conditions during table writes. Therefore, when a table must reject records that violate a defined numeric range, a Delta constraint is an appropriate mechanism.

Question 390

A production pipeline needs to use a package that is not included in the selected Databricks runtime. What should the engineer configure?

1. A watermark
2. A library dependency
3. A row filter
4. A table constraint

Correct Answer: 2

Explanation

A library dependency makes an additional software package available to a Databricks workload when that package is not already included in the runtime environment. Explicitly managing required dependencies helps ensure that jobs have the libraries they need when executed. This is particularly important for production pipelines because missing dependencies can cause runtime failures. A watermark manages event-time state in streaming workloads, a row filter controls which records users can see, and a table constraint enforces data-quality rules. None of these capabilities installs or supplies application libraries. Therefore, when a pipeline requires an additional package, the appropriate library dependency should be configured.

Question 391

A team wants to remove obsolete Delta files while preserving files required for the configured historical retention period. Which operation is designed for this maintenance task?

1. MERGE
2. VACUUM
3. REPARTITION
4. DESCRIBE TABLE

Correct Answer: 2

Explanation

VACUUM is the Delta Lake maintenance operation used to remove obsolete data files that are no longer required according to the applicable retention configuration. Proper retention is important because historical files may be needed for time travel and other recovery or auditing scenarios. Removing files too aggressively can affect the availability of historical versions. MERGE synchronizes records between source and target datasets, REPARTITION changes DataFrame partitioning, and DESCRIBE TABLE provides schema information. Therefore, when the goal is to clean up obsolete Delta files while respecting retention requirements, VACUUM is the appropriate operation.

Question 392

A data engineer wants to query an earlier version of a Delta table to investigate a data issue without permanently changing the current table. Which capability should be used?

1. Delta time travel
2. Cluster autoscaling
3. Job scheduling
4. Storage credential

Correct Answer: 1

Explanation

Delta time travel allows users to access historical versions of a Delta table using supported version or timestamp information. This is useful for investigating data changes, troubleshooting incorrect updates, auditing historical states, and comparing earlier data with the current version. Querying an earlier version does not inherently replace the current table contents. Cluster autoscaling manages compute capacity, job scheduling controls workflow execution times, and storage credentials provide authentication for external storage. These features do not provide historical table querying. Therefore, when an engineer needs to inspect an earlier Delta table state without permanently modifying the current state, Delta time travel is appropriate.

Question 393

A workflow should execute a task every morning at 6:00 AM automatically. Which feature should be configured?

1. Column masking
2. Job schedule
3. Data skipping
4. Unity Catalog volume

Correct Answer: 2

Explanation

A job schedule allows a workflow to execute automatically according to a defined time-based schedule. For a pipeline that must run every morning at a specific time, a scheduled workflow eliminates the need for manual execution. Scheduling is commonly used for recurring ingestion, transformation, reporting, and maintenance workloads. Column masking protects sensitive data, data skipping improves query performance, and Unity Catalog volumes provide governed file storage. None of these features controls when a workflow starts. Therefore, when a job needs to execute automatically at a recurring time such as 6:00 AM each morning, a job schedule is the appropriate workflow capability.

Question 394

A data engineer wants to determine the detailed metadata and storage information for a Delta table before troubleshooting a performance issue. Which command is appropriate?

1. DESCRIBE DETAIL
2. DROP SCHEMA
3. VACUUM
4. MERGE

Correct Answer: 1

Explanation

DESCRIBE DETAIL provides detailed information about a Delta table and can expose useful metadata such as the table location, format, and other properties. This information can help engineers understand how the table is configured before investigating performance or storage-related issues. DROP SCHEMA removes a schema and potentially its contained objects according to the specified behavior, VACUUM removes eligible obsolete files, and MERGE synchronizes source and target records. These operations do not provide the same detailed table metadata. Therefore, when detailed Delta table information is needed for troubleshooting, DESCRIBE DETAIL is the appropriate command.

Question 395

A data pipeline receives records containing unexpected fields that are not part of the currently inferred schema. The engineer wants to preserve those unexpected values instead of discarding them. Which capability can help?

1. Query caching
2. Rescued data column
3. Cluster policy
4. Job retry

Correct Answer: 2

Explanation

A rescued data column can preserve unexpected or incompatible data encountered during supported ingestion processes. This provides a way to retain information that does not fit the expected schema so that engineers can inspect and handle it later instead of simply losing it. This can be particularly useful when source systems evolve or occasionally send inconsistent records. Query caching is a performance feature, cluster policies govern compute configurations, and job retries control repeated workflow execution after failures. None of these features is designed to preserve unexpected input fields. Therefore, when unexpected values need to be retained for investigation, a rescued data column can be useful.

Question 396

A team wants to review proposed changes to a data engineering project before merging them into the main development branch. Which practice should be used?

1. Git pull request
2. VACUUM
3. Data masking
4. Table optimization

Correct Answer: 1

Explanation

A Git pull request provides a structured mechanism for reviewing proposed source-code changes before they are merged into a target branch. Team members can inspect the changes, discuss implementation details, run appropriate validation, and approve or reject the proposed modification according to the team’s development process. This supports collaboration, traceability, and controlled promotion of pipeline code. VACUUM manages obsolete Delta files, data masking protects sensitive values, and table optimization improves data layout. These features do not provide source-code review workflows. Therefore, when a team needs to review proposed project changes before merging them, a Git pull request is an appropriate practice.

Question 397

A data engineer needs to combine a streaming dataset containing transactions with a static customer reference dataset to enrich each transaction. Which operation is appropriate?

1. Stream-static join
2. VACUUM
3. Cluster autoscaling
4. Schema deletion

Correct Answer: 1

Explanation

A stream-static join combines a continuously arriving streaming dataset with a static dataset. This is useful when streaming records need additional information from a reference dataset, such as customer attributes, product descriptions, or organizational mappings. The static reference data can enrich incoming events without itself being processed as a streaming source. VACUUM performs Delta file cleanup, cluster autoscaling adjusts compute capacity, and schema deletion is unrelated to streaming enrichment. Therefore, when transaction events arriving through a stream need to be enriched with information from a static customer dataset, a stream-static join is an appropriate approach.

Question 398

A data engineer wants to reduce the number of small files generated by frequent writes to a Delta table. Which operation should be considered?

1. SHOW GRANTS
2. OPTIMIZE
3. CREATE VIEW
4. DESCRIBE HISTORY

Correct Answer: 2

Explanation

OPTIMIZE can compact small Delta files into larger files and improve the physical organization of table data. Frequent incremental writes can create many small files, which may increase metadata overhead and negatively affect query performance. Running optimization can reduce file fragmentation and make subsequent reads more efficient. SHOW GRANTS is used for permissions, CREATE VIEW creates a logical representation of data, and DESCRIBE HISTORY displays transaction information. These features do not address small-file compaction. Therefore, when frequent writes have resulted in many small Delta files, OPTIMIZE is an appropriate operation to improve the table’s file layout.

Question 399

A data engineer needs a pipeline design where rerunning the same input does not create additional incorrect duplicate records. Which principle should be followed?

1. Idempotent processing
2. Full refresh
3. Manual execution
4. Unrestricted writes

Correct Answer: 1

Explanation

Idempotent processing ensures that processing the same input more than once produces the same intended result rather than accumulating incorrect duplicate effects. This is important in production pipelines because jobs may be retried, restarted, or manually rerun after failures. Techniques such as deterministic keys, appropriate merge logic, checkpointing, and controlled write operations can contribute to an idempotent design. A full refresh repeatedly processes the complete dataset and may be inefficient, manual execution does not address duplicate effects, and unrestricted writes can increase the risk of duplicate data. Therefore, when repeated processing must produce a consistent result without creating incorrect duplicates, idempotent processing is the appropriate principle.

Question 400

A data engineer wants to inspect which users and groups have privileges on a Unity Catalog table while troubleshooting an access issue. Which command should be used?

1. OPTIMIZE
2. VACUUM
3. SHOW GRANTS
4. MERGE

Correct Answer: 3

Explanation

SHOW GRANTS can be used to inspect privileges assigned to supported Unity Catalog objects. When troubleshooting an access issue, reviewing the granted permissions helps determine whether the affected user or group has the required privilege on the table or another relevant object. This makes SHOW GRANTS useful for access troubleshooting and governance verification. OPTIMIZE improves Delta file organization, VACUUM removes eligible obsolete files, and MERGE synchronizes source and target records. These operations do not provide information about table permissions. Therefore, when the objective is to inspect which users or groups have privileges on a Unity Catalog table, SHOW GRANTS is the appropriate command.