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

 

Question 121

Which Unity Catalog privilege is required to read data from a table when the user already has the necessary parent namespace permissions?

1. SELECT
2. MODIFY
3. CREATE TABLE
4. USE SCHEMA

Correct Answer: 1

Explanation

The SELECT privilege allows a user or group to read data from a table when the required parent-level permissions are also available. In Unity Catalog, access is controlled through privileges assigned to securable objects such as catalogs, schemas, tables, and views. A user typically needs the appropriate namespace permissions before table-level SELECT access can be used successfully. MODIFY is associated with changing data, CREATE TABLE permits table creation, and USE SCHEMA provides access to the schema namespace. Therefore, when the requirement is specifically to read rows from an existing table, SELECT is the relevant table-level privilege.

Question 122

Which Azure Databricks feature is designed to provide scalable ingestion of continuously arriving files from cloud storage?

1. Delta Sharing
2. Auto Loader
3. Cluster policy
4. SQL warehouse

Correct Answer: 2

Explanation

Auto Loader provides a scalable approach for incrementally ingesting files that arrive in cloud storage. It is designed to identify new files and process them without repeatedly processing previously discovered files. This makes it useful for data engineering workloads where source systems continuously produce files. Auto Loader can work with Structured Streaming and provides capabilities for schema inference and supported schema evolution scenarios. Delta Sharing is used to share data securely, cluster policies control compute configuration, and SQL warehouses provide SQL query compute. Therefore, Auto Loader is the most appropriate feature when a pipeline needs reliable and scalable ingestion of continuously arriving cloud-storage files.

Question 123

Which Delta Lake capability allows users to inspect or query a table as it existed at an earlier point in time?

1. Schema evolution
2. Auto Loader
3. Time travel
4. Partition pruning

Correct Answer: 3

Explanation

Delta Lake time travel allows users to query previous versions of a Delta table using historical versions or timestamps, provided the required transaction history and data files are still available. This capability is valuable for auditing, debugging, investigating changes, and reproducing historical analytical results. Delta Lake maintains transaction information that records table modifications, making historical versions accessible according to retention and storage conditions. Schema evolution handles structural changes, Auto Loader manages incremental file ingestion, and partition pruning improves query efficiency. Therefore, time travel is the feature specifically designed to provide access to earlier states of Delta table data.

Question 124

Which object provides a governed abstraction for a cloud storage path that can be used with external data in Unity Catalog?

1. Volume
2. External location
3. Materialized view
4. Catalog

Correct Answer: 4

Explanation

An external location represents a governed cloud storage path in Unity Catalog and is associated with a storage credential. It allows administrators to manage access to external storage through centralized Unity Catalog permissions. External locations are commonly used when working with external tables and other data stored outside managed table storage. A volume provides governed access to files, a materialized view stores maintained query results, and a catalog provides a namespace for schemas and objects. The external location specifically connects Unity Catalog governance to a defined cloud storage path, making it the appropriate object when external storage needs centralized authorization and management.

Question 125

Which operation should be used to apply conditional updates and inserts to a Delta table based on matching records?

1. MERGE
2. VACUUM
3. OPTIMIZE
4. DESCRIBE

Correct Answer: 1

Explanation

MERGE is designed for synchronizing data between a source and a target Delta table using defined matching conditions. It can update existing target records when a match is found and insert new records when a match does not exist. Additional conditions can also support delete operations. MERGE is therefore commonly used for incremental loads, change data capture, and maintaining current-state datasets. VACUUM removes obsolete files, OPTIMIZE improves physical file organization, and DESCRIBE returns metadata about an object. When the requirement involves conditional inserts and updates based on a matching key, MERGE provides the appropriate Delta Lake operation.

Question 126

Which Structured Streaming feature records processing progress so that a streaming query can recover after a restart?

1. View
2. Volume
3. Checkpoint
4. Catalog

Correct Answer: 3

Explanation

A checkpoint stores important progress and state information for a Structured Streaming query. When a streaming workload fails or restarts, checkpoint information helps the query continue from its previously recorded processing state rather than treating all input as new. Checkpoints are particularly important for stateful streaming operations and reliable incremental processing. They should be stored at an appropriate persistent location and should remain associated with the intended streaming query. Views, volumes, and catalogs serve different data-management purposes and do not provide streaming recovery state. Therefore, checkpointing is the appropriate mechanism for preserving streaming progress and supporting recovery after failures or restarts.

Question 127

Which architecture separates raw ingestion data from cleaned and business-ready datasets?

1. Star schema
2. Medallion architecture
3. Client-server architecture
4. Hub-and-spoke architecture

Correct Answer: 2

Explanation

The medallion architecture organizes data into progressively refined layers, commonly called Bronze, Silver, and Gold. Bronze generally contains raw or minimally processed source data. Silver applies transformations such as cleansing, validation, deduplication, and standardization. Gold contains curated data prepared for business analytics, reporting, and downstream consumption. This layered structure provides a clear separation between ingestion, transformation, and business-facing data products. A star schema is a dimensional modeling pattern, while client-server and hub-and-spoke architectures address different application or organizational designs. Therefore, the medallion architecture is the appropriate pattern when a data platform needs separate raw, cleaned, and business-ready layers.

Question 128

Which Unity Catalog privilege provides access to the namespace of a catalog?

1. SELECT
2. MODIFY
3. USE CATALOG
4. CREATE TABLE

Correct Answer: 3

Explanation

USE CATALOG allows a principal to access a catalog namespace in Unity Catalog. It is typically required before users can work with schemas and objects contained within that catalog, along with the necessary permissions at lower levels. SELECT grants read access to supported data objects, MODIFY allows permitted data changes, and CREATE TABLE controls the ability to create tables when the required parent permissions are available. Unity Catalog uses hierarchical access control, so permissions at one level do not automatically replace permissions required at another level. Therefore, USE CATALOG is the relevant privilege when the requirement is access to the catalog namespace.

Question 129

Which feature is most appropriate for masking sensitive values in a column based on the identity or permissions of the user?

1. Column masking
2. Partition pruning
3. File compaction
4. Schema evolution

Correct Answer: 4

Explanation

Column masking is used to control how sensitive column values are presented to users. A masking policy can return the original value for authorized users while presenting a transformed or obscured value to users who should not see the sensitive information directly. This supports fine-grained governance while allowing the same table to remain available for broader analytical use. Partition pruning and file compaction are performance-related techniques, while schema evolution manages changes to data structures. Column masking directly addresses the visibility of sensitive field values, making it the appropriate feature when data access needs to vary according to user identity or authorization.

Question 130

Which operation can reduce the number of small files in a Delta table?

1. GRANT
2. DESCRIBE
3. CREATE VIEW
4. OPTIMIZE

Correct Answer: 4

Explanation

OPTIMIZE can reorganize Delta table files and compact many small files into fewer larger files. Small files can be created by frequent incremental writes, streaming workloads, or numerous small batch operations. Over time, a large number of small files can increase file-management overhead and negatively affect query performance. OPTIMIZE addresses this physical storage-layout problem without changing the logical meaning of the table’s data. GRANT manages permissions, DESCRIBE provides metadata, and CREATE VIEW defines reusable query logic. Therefore, OPTIMIZE is the appropriate Delta Lake maintenance operation when the primary concern is reducing small-file overhead and improving physical data organization.

Question 131

Which feature can help a streaming pipeline handle events that arrive later than their event timestamps?

1. Watermarking
2. Git folders
3. External locations
4. Materialized views

Correct Answer: 1

Explanation

Watermarking helps Structured Streaming manage late-arriving events by establishing an event-time threshold for state management. It allows the streaming engine to determine how long it should retain state while waiting for delayed records. This is especially useful for stateful operations such as aggregations, joins, and deduplication. Once the watermark advances beyond an appropriate threshold, older state can eventually be removed, helping control resource consumption. Git folders support source control, external locations govern cloud storage paths, and materialized views provide maintained query results. Therefore, watermarking is the relevant Structured Streaming capability when a pipeline must account for events that arrive after their expected event time.

Question 132

Which technique is most appropriate for ensuring that rerunning a failed data pipeline does not create duplicate target records?

1. Disable checkpoints
2. Use idempotent processing
3. Delete all previous data
4. Increase worker memory

Correct Answer: 2

Explanation

Idempotent processing allows a pipeline to be safely executed multiple times without producing unintended duplicate effects. This is particularly important when jobs are retried after failures or when source systems can deliver the same data more than once. Stable business keys, deduplication logic, MERGE operations, and appropriate checkpointing can all contribute to an idempotent design. Disabling checkpoints can reduce streaming recovery capabilities, deleting previous data is unsafe and unnecessary, and increasing worker memory addresses resource capacity rather than duplicate prevention. Therefore, implementing idempotent processing logic is the appropriate approach for making a pipeline safe to rerun while maintaining correct target data.

Question 133

Which Databricks capability can orchestrate multiple data engineering tasks with dependencies, schedules, and retry settings?

1. Unity Catalog
2. Delta Sharing
3. Lakeflow Jobs
4. External location

Correct Answer: 3

Explanation

Lakeflow Jobs, the current Databricks terminology for job orchestration capabilities, can coordinate multiple tasks within an automated workflow. A workflow can define task dependencies, schedules or triggers, retry behavior, notifications, and other execution settings. This allows data engineers to build repeatable pipelines where tasks execute in the correct order and failures can be handled according to configured policies. Unity Catalog provides governance, Delta Sharing provides data-sharing capabilities, and external locations govern cloud storage paths. Therefore, Lakeflow Jobs is the appropriate capability for orchestrating data engineering workloads that contain multiple dependent tasks and operational requirements.

Question 134

Which technique can reduce unnecessary data scanning when a query filters on a partition column?

1. Schema evolution
2. Time travel
3. Partition pruning
4. Column masking

Correct Answer: 3

Explanation

Partition pruning allows the query engine to skip partitions that cannot contain records matching a filter condition. For example, a table partitioned by date can avoid scanning unrelated dates when a query requests a specific date range. This reduces the amount of data that must be read and processed and can improve query performance. Partition pruning is a physical data-layout optimization and works most effectively when queries contain filters that align with partitioning columns. Schema evolution manages structural changes, time travel provides access to historical Delta versions, and column masking controls sensitive-value visibility. Therefore, partition pruning is the appropriate technique for reducing unnecessary partition scans.

Question 135

Which Unity Catalog object is designed to provide governed access to non-tabular files?

1. Volume
2. Catalog
3. Schema
4. Table

Correct Answer: 1

Explanation

A volume provides governed file access through Unity Catalog and is intended for file-based data that may not naturally be represented as relational tables. Examples include CSV files, JSON documents, images, machine-learning artifacts, and other unstructured or semi-structured files. Volumes allow organizations to apply centralized permissions and governance to these files. A catalog groups schemas, a schema organizes objects, and a table represents structured tabular data. Therefore, when a data engineer needs governed access to non-tabular files within the Unity Catalog framework, a volume is the appropriate object to use.

Question 136

Which Delta Lake operation removes old files that are no longer required according to the configured retention period?

1. MERGE
2. VACUUM
3. OPTIMIZE
4. SELECT

Correct Answer: 2

Explanation

VACUUM removes obsolete data files from a Delta table after they are no longer needed according to the applicable retention configuration. It is a storage-maintenance operation that helps control the amount of unused data stored in cloud storage. VACUUM should be used carefully because removing files can affect access to historical table versions that depend on those files. MERGE changes table records, OPTIMIZE reorganizes and compacts files, and SELECT reads data. Therefore, VACUUM is the operation specifically intended to clean up obsolete Delta data files while respecting the table’s retention requirements.

Question 137

Which feature allows an organization to control permitted settings when users create Databricks compute resources?

1. Time travel
2. Cluster policy
3. Auto Loader
4. Materialized view

Correct Answer: 4

Explanation

Cluster policies allow administrators to define and restrict compute configuration settings for Databricks resources. Policies can standardize or limit settings such as runtime versions, node types, autoscaling ranges, and other supported parameters. This helps organizations maintain consistent compute configurations and enforce operational or governance requirements. Time travel is a Delta Lake historical-data feature, Auto Loader handles incremental file ingestion, and materialized views provide maintained query results. Cluster policies therefore address the governance and standardization of compute configuration rather than data ingestion or storage. They are especially useful when administrators need to control how users provision shared or job-related compute resources.

Question 138

Which operation can be used to redistribute records across Spark partitions before subsequent processing?

1. CACHE
2. REPARTITION
3. GRANT
4. VACUUM

Correct Answer: 2

Explanation

REPARTITION redistributes records across Spark partitions and can change the number of partitions used by a dataset. It can be useful when the existing partition distribution is unsuitable for the workload or when greater parallelism is required. Because repartitioning commonly causes a shuffle, it can introduce additional network and processing overhead, so it should be used based on actual workload requirements. CACHE stores computed results for reuse, GRANT manages permissions, and VACUUM removes obsolete Delta files. Therefore, REPARTITION is the appropriate Spark operation when the goal is to redistribute data across partitions for subsequent processing.

Question 139

Which feature allows a data engineer to apply custom processing logic separately to each micro-batch in a Structured Streaming query?

1. foreachBatch
2. USE SCHEMA
3. OPTIMIZE
4. SHOW GRANTS

Correct Answer: 1

Explanation

foreachBatch allows custom processing logic to be applied to each micro-batch generated by a Structured Streaming query. This can be useful when streaming data needs to be written using custom logic or when a batch-oriented operation, such as a Delta MERGE, needs to be performed for each micro-batch. The approach provides flexibility while still allowing the source workload to operate as a streaming query. USE SCHEMA changes the active schema context, OPTIMIZE manages Delta file organization, and SHOW GRANTS displays permissions. Therefore, foreachBatch is the appropriate feature when custom processing needs to be executed separately for every micro-batch.

Question 140

Which capability provides a centralized view of permissions assigned to users and groups for Unity Catalog objects?

1. Auto Loader
2. Spark UI
3. SHOW GRANTS
4. OPTIMIZE

Correct Answer: 3

Explanation

SHOW GRANTS can be used to display permissions and grants associated with Unity Catalog securable objects. This is useful for administrators and data engineers who need to inspect which users, groups, or service principals have been assigned specific privileges. Reviewing grants can help troubleshoot access problems and verify that governance policies have been applied as intended. Auto Loader handles incremental ingestion, Spark UI is used for monitoring Spark execution, and OPTIMIZE improves Delta table file organization. Therefore, SHOW GRANTS is the appropriate capability when the requirement is to inspect permissions assigned to principals for governed Unity Catalog objects.