Microsoft DP-900 Practice Test 31
Question 300
A scheduled dataset refresh must retry if one source connection times out. What is the MOST appropriate next validation or correction?
- Validate measure filter context and refresh timestamp.
- Simulate upstream failures and inspect pipeline run dependencies.
- Validate measure totals across two consuming reports.
- Set three tasks to 01:00 and claim they automatically wait for one another.
Explanation
Correct Answer
B:Simulate upstream failures and inspect pipeline run dependencies. This tests the intended result: dependency stages run in the intended order and failures are visible.
Incorrect Answers
A:Validate measure filter context and refresh timestamp. It tests KPI card single metric rather than the critical Fabric Data Factory pipeline orchestration behavior. Orchestration manages data movement and execution dependencies.
C:Validate measure totals across two consuming reports. It tests Semantic model over curated analytics rather than the critical Fabric Data Factory pipeline orchestration behavior. Orchestration manages data movement and execution dependencies.
D:Set three tasks to 01:00 and claim they automatically wait for one another. This unsafe shortcut leaves the Fabric Data Factory pipeline orchestration problem unresolved. Starting every activity simultaneously may publish partial outputs.
Question 301
A source API changes JSON field names unexpectedly. After selecting the proper Data engineer schema and format changes approach, which observation gives the clearest evidence of success?
- Rows containing quoted delimiters retain the intended field count.
- Analytics engines list intended folders and permissions govern directory access.
- Unexpected schema changes are quarantined or mapped deliberately.
- Tests confirm analysts can SELECT approved data but cannot DROP tables.
Explanation
Correct Answer
C:Unexpected schema changes are quarantined or mapped deliberately. This is the directly observable result for Data engineer schema and format changes; data engineers maintain resilient transformations when source structures change.
Incorrect Answers
A:Rows containing quoted delimiters retain the intended field count. This supports CSV delimited text schema, not confirmation of Data engineer schema and format changes. The relevant evidence is that unexpected schema changes are quarantined or mapped deliberately.
B:Analytics engines list intended folders and permissions govern directory access. This supports Azure Data Lake Storage Gen2 analytics landing, not confirmation of Data engineer schema and format changes. The relevant evidence is that unexpected schema changes are quarantined or mapped deliberately.
D:Tests confirm analysts can SELECT approved data but cannot DROP tables. This supports DBA least privilege account roles, not confirmation of Data engineer schema and format changes. The relevant evidence is that unexpected schema changes are quarantined or mapped deliberately.
Question 302
An accounting team needs strongly modeled fact and dimension tables using SQL. If the relevant Fabric Warehouse relational reporting safeguard is skipped, which failure is MOST directly associated with that omission?
- Running hourly batch for urgent alerts creates unacceptable delay.
- Advancing the watermark before successful commit may skip data.
- Using a warehouse as a replacement for every OLTP payment transaction may be inappropriate.
- Assuming Import updates automatically for every new source row leads to stale insights.
Explanation
Correct Answer
C:Using a warehouse as a replacement for every OLTP payment transaction may be inappropriate. Fabric Warehouse provides a SQL-oriented analytics experience in Fabric. A suitable check is to test analytical SELECT and supported load behavior.
Incorrect Answers
A:Running hourly batch for urgent alerts creates unacceptable delay. This failure is associated with Streaming versus batch latency choice; it is not the primary failure in the stated Fabric Warehouse relational reporting scenario. Fabric Warehouse provides a SQL-oriented analytics experience in Fabric.
B:Advancing the watermark before successful commit may skip data. This failure is associated with Incremental ingestion with watermark; it is not the primary failure in the stated Fabric Warehouse relational reporting scenario. Fabric Warehouse provides a SQL-oriented analytics experience in Fabric.
D:Assuming Import updates automatically for every new source row leads to stale insights. This failure is associated with Power BI Import versus DirectQuery mode; it is not the primary failure in the stated Fabric Warehouse relational reporting scenario. Fabric Warehouse provides a SQL-oriented analytics experience in Fabric.
Question 303
An order system must reject negative line quantities and missing currency. If the relevant Relational constraint CHECK and NOT NULL safeguard is skipped, which failure is MOST directly associated with that omission?
- Ignoring transient failover errors can disrupt applications.
- UI-only validation can be bypassed by other write clients.
- Copying the customer name into every order risks inconsistent updates.
- Granting public network access broadly increases exposure regardless of SQL login strength.
Explanation
Correct Answer
B:UI-only validation can be bypassed by other write clients. Column and row constraints enforce business validity beyond application forms. A suitable check is to attempt valid invalid and null inserts with test transactions.
Incorrect Answers
A:Ignoring transient failover errors can disrupt applications. This failure is associated with Azure SQL high availability and zone resiliency; it is not the primary failure in the stated Relational constraint CHECK and NOT NULL scenario. Column and row constraints enforce business validity beyond application forms.
C:Copying the customer name into every order risks inconsistent updates. This failure is associated with One to many customer order relationship; it is not the primary failure in the stated Relational constraint CHECK and NOT NULL scenario. Column and row constraints enforce business validity beyond application forms.
D:Granting public network access broadly increases exposure regardless of SQL login strength. This failure is associated with Azure SQL firewall and private endpoint; it is not the primary failure in the stated Relational constraint CHECK and NOT NULL scenario. Column and row constraints enforce business validity beyond application forms.
Question 304
An e-commerce service retrieves shopping-cart state by user ID. Which principle BEST explains the correct Cosmos DB NoSQL JSON document container design decision?
- Page blobs are suited to random I/O scenarios and VHD backing.
- Multiple write regions add concurrency and conflict resolution concerns.
- The NoSQL API stores JSON items without requiring identical SQL rows.
- Soft delete retains recoverable deleted content for a configured period.
Explanation
Correct Answer
C:The NoSQL API stores JSON items without requiring identical SQL rows. In this case the applicable decision is to use Azure Cosmos DB for NoSQL with suitable documents and container partition strategy.
Incorrect Answers
A:Page blobs are suited to random I/O scenarios and VHD backing. That concept explains Page blob random access; the situation instead concerns Cosmos DB NoSQL JSON document container. Selecting a relational JOIN-heavy schema without model review may inflate request cost.
B:Multiple write regions add concurrency and conflict resolution concerns. That concept explains Cosmos DB multi-region writes conflict handling; the situation instead concerns Cosmos DB NoSQL JSON document container. Selecting a relational JOIN-heavy schema without model review may inflate request cost.
D:Soft delete retains recoverable deleted content for a configured period. That concept explains Blob soft delete recovery; the situation instead concerns Cosmos DB NoSQL JSON document container. Selecting a relational JOIN-heavy schema without model review may inflate request cost.
Question 305
A development database is idle overnight and receives bursty daytime requests. If the relevant Azure SQL Database serverless compute safeguard is skipped, which failure is MOST directly associated with that omission?
- Blindly moving instance-specific code to Azure SQL Database can fail.
- Embedding lists of related IDs in one column makes constraints and joins harder.
- Assuming zero start latency after auto-pause can disappoint clients.
- Missing relationships may disappear from inner-join results.
Explanation
Correct Answer
C:Assuming zero start latency after auto-pause can disappoint clients. Serverless compute can adjust resources and potentially pause eligible idle databases. A suitable check is to test idle periods resume delays and minimum billing constraints.
Incorrect Answers
A:Blindly moving instance-specific code to Azure SQL Database can fail. This failure is associated with Azure SQL Managed Instance compatibility; it is not the primary failure in the stated Azure SQL Database serverless compute scenario. Serverless compute can adjust resources and potentially pause eligible idle databases.
B:Embedding lists of related IDs in one column makes constraints and joins harder. This failure is associated with Many to many bridge table; it is not the primary failure in the stated Azure SQL Database serverless compute scenario. Serverless compute can adjust resources and potentially pause eligible idle databases.
D:Missing relationships may disappear from inner-join results. This failure is associated with Inner join row matching; it is not the primary failure in the stated Azure SQL Database serverless compute scenario. Serverless compute can adjust resources and potentially pause eligible idle databases.
Question 306
An accounting team needs strongly modeled fact and dimension tables using SQL. Which choice BEST meets this requirement?
- Call a Lakehouse file browser a SQL Warehouse without creating its relational model.
- Use inactivity-based session windows to group related event sequences.
- Use a Fabric Warehouse for supported SQL-centric analytical table operations.
- Ingest raw data then perform controlled ELT transformations within target compute.
Explanation
Correct Answer
C:Fabric Warehouse provides a SQL-oriented analytics experience in Fabric.
Incorrect Answers
A:Call a Lakehouse file browser a SQL Warehouse without creating its relational model. It does not implement Fabric Warehouse relational reporting. Using a warehouse as a replacement for every OLTP payment transaction may be inappropriate.
B:Use inactivity-based session windows to group related event sequences. That action primarily handles Session windows of event activity; the given requirement is Fabric Warehouse relational reporting. Fabric Warehouse provides a SQL-oriented analytics experience in Fabric.
D:Ingest raw data then perform controlled ELT transformations within target compute. That action primarily handles ELT warehouse pushdown transforms; the given requirement is Fabric Warehouse relational reporting. Fabric Warehouse provides a SQL-oriented analytics experience in Fabric.
Question 307
A small team wants managed MySQL hosting instead of patching a VM database. What is the MOST appropriate next validation or correction?
- Profile representative transaction latency and historical scan plans.
- Replay old and new sample payloads through tests.
- Assume Azure SQL Managed Instance runs the MySQL protocol because both use SQL.
- Test representative stored statements client connectivity and sizing.
Explanation
Correct Answer
D:Test representative stored statements client connectivity and sizing. This tests the intended result: existing MySQL queries and connections work with a managed service.
Incorrect Answers
A:Profile representative transaction latency and historical scan plans. It tests OLTP relational versus OLAP analytical store rather than the critical Azure MySQL versus PostgreSQL service-model selection behavior. Managed open-source database services retain engine-specific behavior.
B:Replay old and new sample payloads through tests. It tests Data engineer schema and format changes rather than the critical Azure MySQL versus PostgreSQL service-model selection behavior. Managed open-source database services retain engine-specific behavior.
C:Assume Azure SQL Managed Instance runs the MySQL protocol because both use SQL. This unsafe shortcut leaves the Azure MySQL versus PostgreSQL service-model selection problem unresolved. Moving directly to a different engine without migration assessment risks SQL incompatibility.
Question 308
Client logic retries instantly in a tight loop during RU saturation. What is the MOST appropriate next validation or correction?
- Simulate throttling and inspect retry behavior.
- Test multi-edge routes and direction filters on sample graph.
- Test object retrieval SMB mounting and key lookups separately.
- Treat every 429 as proof that the user password is invalid.
Explanation
Correct Answer
A:Simulate throttling and inspect retry behavior. This tests the intended result: throttled operations eventually succeed without uncontrolled retry storms.
Incorrect Answers
B:Test multi-edge routes and direction filters on sample graph. It tests Cosmos DB API for Apache Gremlin graph rather than the critical Cosmos DB 429 throttling retries behavior. Request-rate limits protect provisioned throughput and require graceful client handling.
C:Test object retrieval SMB mounting and key lookups separately. It tests Choosing Blob Files and Table for app needs rather than the critical Cosmos DB 429 throttling retries behavior. Request-rate limits protect provisioned throughput and require graceful client handling.
D:Treat every 429 as proof that the user password is invalid. This unsafe shortcut leaves the Cosmos DB 429 throttling retries problem unresolved. Immediate unbounded retries can increase pressure and latency.
Question 309
A team expects a partition-key lookup to find every synonym across product text. What is the MOST appropriate next validation or correction?
- Test restore points separately from zone failover behaviors.
- Use a customer ID hash as a natural-language semantic ranking index.
- Compare exact-ID lookups with text-search queries and relevance.
- Check cohort data and alternative explanations for the decline.
Explanation
Correct Answer
C:Compare exact-ID lookups with text-search queries and relevance. This tests the intended result: searches find relevant content while record reads remain reliable.
Incorrect Answers
A:Test restore points separately from zone failover behaviors. It tests Storage resiliency and replicas rather than the critical Document indexing versus full-text search behavior. Point lookups and lexical text retrieval use different indexing models.
B:Use a customer ID hash as a natural-language semantic ranking index. This unsafe shortcut leaves the Document indexing versus full-text search problem unresolved. Assuming every document database query automatically provides relevance ranking can disappoint users.
D:Check cohort data and alternative explanations for the decline. It tests Descriptive versus diagnostic analytics rather than the critical Document indexing versus full-text search behavior. Point lookups and lexical text retrieval use different indexing models.
Question 310
An inventory edit fails validation before the purchase completes. What is the MOST appropriate next validation or correction?
- Inject an error between updates and compare resulting balances.
- Treat a dashboard refresh as SQL transaction COMMIT.
- Follow functional dependencies and test a rename operation.
- Exercise retry logic and inspect tier redundancy capabilities.
Explanation
Correct Answer
A:Inject an error between updates and compare resulting balances. This tests the intended result: both sides are committed together or no partial change remains.
Incorrect Answers
B:Treat a dashboard refresh as SQL transaction COMMIT. This unsafe shortcut leaves the SQL COMMIT and ROLLBACK transaction problem unresolved. Independent nontransactional updates can leave inconsistent totals.
C:Follow functional dependencies and test a rename operation. It tests Third normal form transitive dependency rather than the critical SQL COMMIT and ROLLBACK transaction behavior. Atomicity is an ACID property of relational transactions.
D:Exercise retry logic and inspect tier redundancy capabilities. It tests Azure SQL high availability and zone resiliency rather than the critical SQL COMMIT and ROLLBACK transaction behavior. Atomicity is an ACID property of relational transactions.