Microsoft DP-900 Practice Test 64
Question 630
A service mostly writes individual inventory changes while another scans monthly patterns. After selecting the proper Data platform read and write profiles approach, which observation gives the clearest evidence of success?
- Workload tests reflect transaction speed and analytical scans independently.
- Aggregates group actual status codes rather than arbitrary matching text.
- Checkout remains responsive while historical scans execute on suitable analytical storage.
- Tests confirm analysts can SELECT approved data but cannot DROP tables.
Explanation
Correct Answer
A:Workload tests reflect transaction speed and analytical scans independently. This is the directly observable result for Data platform read and write profiles; data store suitability depends on workload access patterns not only data types.
Incorrect Answers
B:Aggregates group actual status codes rather than arbitrary matching text. This supports Log text extraction versus structured event fields, not confirmation of Data platform read and write profiles. The relevant evidence is that workload tests reflect transaction speed and analytical scans independently.
C:Checkout remains responsive while historical scans execute on suitable analytical storage. This supports OLTP relational versus OLAP analytical store, not confirmation of Data platform read and write profiles. The relevant evidence is that workload tests reflect transaction speed and analytical scans independently.
D:Tests confirm analysts can SELECT approved data but cannot DROP tables. This supports DBA least privilege account roles, not confirmation of Data platform read and write profiles. The relevant evidence is that workload tests reflect transaction speed and analytical scans independently.
Question 631
Stakeholders cannot interpret an overcomplicated revenue dashboard. During testing, the team observes this specific failure: Choosing a misleading graphic can hide a real business pattern. Which corrective check should be performed FIRST?
- Choose Azure SQL Database PaaS with appropriate compute and resilience settings.
- Test dashboard comprehension using business questions.
- Assign database administration responsibility for backup retention restore testing and database health.
- Call a missing data ingestion job complete because the chart looks attractive.
Explanation
Correct Answer
B:Test dashboard comprehension using business questions. A successful correction should demonstrate that users can answer agreed questions from accurate accessible visuals.
Incorrect Answers
A:Choose Azure SQL Database PaaS with appropriate compute and resilience settings. It serves Azure SQL Database managed relational app, not the diagnosed Data analyst report design problem. Choosing a misleading graphic can hide a real business pattern.
C:Assign database administration responsibility for backup retention restore testing and database health. It serves DBA backup and recovery responsibility, not the diagnosed Data analyst report design problem. Choosing a misleading graphic can hide a real business pattern.
D:Call a missing data ingestion job complete because the chart looks attractive. This reinforces the error rather than addressing the Data analyst report design failure. Analysts translate curated data into interpretable insights.
Question 632
A service needs a direct lookup by device key without SQL joins. Which principle BEST explains the correct Table Storage PartitionKey and RowKey design decision?
- Change feeds support incremental change-driven processing with defined change semantics.
- Table Storage identifies entities by partition and row keys.
- Similar keyword syntax does not make NoSQL document storage a SQL Server database.
- Block blobs suit file and streaming upload scenarios with manageable blocks.
Explanation
Correct Answer
B:Table Storage identifies entities by partition and row keys. In this case the applicable decision is to design entities with appropriate PartitionKey and RowKey values.
Incorrect Answers
A:Change feeds support incremental change-driven processing with defined change semantics. That concept explains Cosmos DB change feed updates; the situation instead concerns Table Storage PartitionKey and RowKey. Unbalanced partition keys or unselective scans can harm performance.
C:Similar keyword syntax does not make NoSQL document storage a SQL Server database. That concept explains Cosmos DB API for NoSQL versus SQL text; the situation instead concerns Table Storage PartitionKey and RowKey. Unbalanced partition keys or unselective scans can harm performance.
D:Block blobs suit file and streaming upload scenarios with manageable blocks. That concept explains Block blob file uploads; the situation instead concerns Table Storage PartitionKey and RowKey. Unbalanced partition keys or unselective scans can harm performance.
Question 633
A website uploads large PDF and video assets in chunks. After selecting the proper Block blob file uploads approach, which observation gives the clearest evidence of success?
- Content assembles correctly and clients retrieve the intended object version.
- Folder-level operations and permissions work as designed.
- Billing and throttling behavior match measured usage and service limits.
- Downstream reflects expected eligible changes without repeated full scans.
Explanation
Correct Answer
A:Content assembles correctly and clients retrieve the intended object version. This is the directly observable result for Block blob file uploads; block blobs suit file and streaming upload scenarios with manageable blocks.
Incorrect Answers
B:Folder-level operations and permissions work as designed. This supports ADLS Gen2 hierarchical namespace, not confirmation of Block blob file uploads. The relevant evidence is that content assembles correctly and clients retrieve the intended object version.
C:Billing and throttling behavior match measured usage and service limits. This supports Cosmos DB serverless versus provisioned throughput, not confirmation of Block blob file uploads. The relevant evidence is that content assembles correctly and clients retrieve the intended object version.
D:Downstream reflects expected eligible changes without repeated full scans. This supports Cosmos DB change feed updates, not confirmation of Block blob file uploads. The relevant evidence is that content assembles correctly and clients retrieve the intended object version.
Question 634
A legacy MySQL app needs managed hosting with minimal dialect changes. Which principle BEST explains the correct Azure Database for MySQL Flexible Server design decision?
- Azure hosts distinct managed open-source SQL engines with their own syntax.
- Atomicity is an ACID property of relational transactions.
- Read scale can separate query consumption from primary writes but may have replication lag.
- Second normal form removes partial dependencies on a composite key.
Explanation
Correct Answer
A:Azure hosts distinct managed open-source SQL engines with their own syntax. In this case the applicable decision is to choose Azure Database for MySQL Flexible Server if version and feature requirements fit.
Incorrect Answers
B:Atomicity is an ACID property of relational transactions. That concept explains SQL COMMIT and ROLLBACK transaction; the situation instead concerns Azure Database for MySQL Flexible Server. Migrating from MySQL to Azure SQL Database untested can change behavior.
C:Read scale can separate query consumption from primary writes but may have replication lag. That concept explains Azure SQL read scale and reporting replicas; the situation instead concerns Azure Database for MySQL Flexible Server. Migrating from MySQL to Azure SQL Database untested can change behavior.
D:Second normal form removes partial dependencies on a composite key. That concept explains Second normal form partial dependency; the situation instead concerns Azure Database for MySQL Flexible Server. Migrating from MySQL to Azure SQL Database untested can change behavior.
Question 635
A team needs to identify outliers in price versus unit demand. What is the MOST appropriate next validation or correction?
- Verify each point's grain and inspect outliers.
- Reconcile each source export with landed records and failure logs.
- Use alphabetical store names for both numeric axes and call it a correlation plot.
- Validate source connection types and loaded sample values.
Explanation
Correct Answer
A:Verify each point's grain and inspect outliers. This tests the intended result: points reveal clustering and possible outliers without asserting causality.
Incorrect Answers
B:Reconcile each source export with landed records and failure logs. It tests Analytical ingestion from operational systems rather than the critical Scatter chart relationship behavior. Scatter charts show relationships between two quantitative variables.
C:Use alphabetical store names for both numeric axes and call it a correlation plot. This unsafe shortcut leaves the Scatter chart relationship problem unresolved. Correlation in points alone does not prove one variable caused the other.
D:Validate source connection types and loaded sample values. It tests Power BI Desktop data import rather than the critical Scatter chart relationship behavior. Scatter charts show relationships between two quantitative variables.
Question 636
A prototype has sporadic low-volume document requests. After selecting the proper Cosmos DB serverless versus provisioned throughput approach, which observation gives the clearest evidence of success?
- Intended queries work with measurable RU use under representative data.
- Billing and throttling behavior match measured usage and service limits.
- Recovery expectations match documented replica placement and failover support.
- New and old devices can be read without demanding identical property sets.
Explanation
Correct Answer
B:Billing and throttling behavior match measured usage and service limits. This is the directly observable result for Cosmos DB serverless versus provisioned throughput; cosmos DB consumption and reserved throughput models suit different demand patterns.
Incorrect Answers
A:Intended queries work with measurable RU use under representative data. This supports Cosmos DB NoSQL indexing and queries, not confirmation of Cosmos DB serverless versus provisioned throughput. The relevant evidence is that billing and throttling behavior match measured usage and service limits.
C:Recovery expectations match documented replica placement and failover support. This supports Azure Storage LRS ZRS GRS distinctions, not confirmation of Cosmos DB serverless versus provisioned throughput. The relevant evidence is that billing and throttling behavior match measured usage and service limits.
D:New and old devices can be read without demanding identical property sets. This supports Table Storage schemaless properties, not confirmation of Cosmos DB serverless versus provisioned throughput. The relevant evidence is that billing and throttling behavior match measured usage and service limits.
Question 637
A contact record has no KnownMiddleName value. If the relevant NULL and three-valued logic safeguard is skipped, which failure is MOST directly associated with that omission?
- Comparing = NULL can silently yield no rows.
- Selecting PaaS for unsupported low-level dependencies breaks the app.
- Treating a VM as fully managed PaaS can leave guest patching unattended.
- Casual duplication without refresh governance makes reports disagree.
Explanation
Correct Answer
A:Comparing = NULL can silently yield no rows. SQL NULL represents absent or unknown values and ordinary equality does not match it. A suitable check is to test true false and unknown predicate cases.
Incorrect Answers
B:Selecting PaaS for unsupported low-level dependencies breaks the app. This failure is associated with SQL deployment choice by control plane; it is not the primary failure in the stated NULL and three-valued logic scenario. SQL NULL represents absent or unknown values and ordinary equality does not match it.
C:This failure is associated with SQL Server on Azure VM OS control; it is not the primary failure in the stated NULL and three-valued logic scenario.
D:Casual duplication without refresh governance makes reports disagree. This failure is associated with Denormalization for analytics tradeoff; it is not the primary failure in the stated NULL and three-valued logic scenario. SQL NULL represents absent or unknown values and ordinary equality does not match it.
Question 638
An import displays corrupted accents and misreads commas used as decimal separators. Which principle BEST explains the correct Text encoding and numeric parsing design decision?
- Analysts must interpret findings with data quality and context.
- Replication improves availability but is not always a replacement for recoverable historic versions.
- Structured relational and semi-structured document models solve different shape and access needs.
- Data representation includes encoding and type interpretation beyond the filename.
Explanation
Correct Answer
D:Data representation includes encoding and type interpretation beyond the filename. In this case the applicable decision is to choose explicit text encoding and locale-aware numeric conversion rules.
Incorrect Answers
A:Analysts must interpret findings with data quality and context. That concept explains Data analyst anomaly interpretation; the situation instead concerns Text encoding and numeric parsing. Treating localized decimals as strings can silently corrupt arithmetic.
B:Replication improves availability but is not always a replacement for recoverable historic versions. That concept explains Storage resiliency and replicas; the situation instead concerns Text encoding and numeric parsing. Treating localized decimals as strings can silently corrupt arithmetic.
C:Structured relational and semi-structured document models solve different shape and access needs. That concept explains Relational rows versus document attributes; the situation instead concerns Text encoding and numeric parsing. Treating localized decimals as strings can silently corrupt arithmetic.
Question 639
Incoming equipment events must be filtered and aggregated every minute. After selecting the proper Azure Stream Analytics SQL-like streaming approach, which observation gives the clearest evidence of success?
- A single event can contribute to overlapping eligible windows.
- Locations resolve to intended regions and values match actual places.
- Output windows contain correct event aggregates under steady ingestion.
- Report filters produce consistent business calculations.
Explanation
Correct Answer
C:Output windows contain correct event aggregates under steady ingestion. This is the directly observable result for Azure Stream Analytics SQL-like streaming; stream Analytics analyzes streams using continuous SQL-like queries.
Incorrect Answers
A:A single event can contribute to overlapping eligible windows. This supports Hopping sliding analytic windows, not confirmation of Azure Stream Analytics SQL-like streaming. The relevant evidence is that output windows contain correct event aggregates under steady ingestion.
B:Locations resolve to intended regions and values match actual places. This supports Map visual geographic measures, not confirmation of Azure Stream Analytics SQL-like streaming. The relevant evidence is that output windows contain correct event aggregates under steady ingestion.
D:Report filters produce consistent business calculations. This supports Power BI semantic model measures, not confirmation of Azure Stream Analytics SQL-like streaming. The relevant evidence is that output windows contain correct event aggregates under steady ingestion.
Question 640
An IoT system must govern which physical device can send event readings. Which principle BEST explains the correct Azure IoT Hub device messaging design decision?
- Notebooks can combine code documentation and analytical results but require governed execution.
- Hopping windows allow overlapping periodic aggregation.
- Fabric Warehouse provides a SQL-oriented analytics experience in Fabric.
- IoT Hub adds device management and identities around IoT messages.
Explanation
Correct Answer
D:IoT Hub adds device management and identities around IoT messages. In this case the applicable decision is to choose Azure IoT Hub for supported device-oriented connectivity and management.
Incorrect Answers
A:Notebooks can combine code documentation and analytical results but require governed execution. That concept explains Databricks collaborative notebooks; the situation instead concerns Azure IoT Hub device messaging. A generic anonymous event endpoint may not satisfy per-device identity policies.
B:Hopping windows allow overlapping periodic aggregation. That concept explains Hopping sliding analytic windows; the situation instead concerns Azure IoT Hub device messaging. A generic anonymous event endpoint may not satisfy per-device identity policies.
C:Fabric Warehouse provides a SQL-oriented analytics experience in Fabric. That concept explains Fabric Warehouse relational reporting; the situation instead concerns Azure IoT Hub device messaging. A generic anonymous event endpoint may not satisfy per-device identity policies.
