Microsoft DP-800 Practice Test 58 - Prepaway
Practice Exams:

Pass Microsoft DP-800 Exam in First Attempt Guaranteed!

Get 100% Latest Exam Questions, Accurate & Verified Answers!

30 Days Free Updates, Instant Download!

Verified By Experts
Premium File
$69.99 $76.99

DP-800 Premium ETE File


  • Premium File 101 Questions & Answers. Last Update: Nov 06, 2026
Whats Included:
  • Latest Questions
  • 100% Accurate Answers
  • Fast Exam Updates

Microsoft DP-800 Practice Test 58

Question 570

A multitenant database serves a tiny customer and a giant enterprise through the same stored procedure; one cached plan is poor for the other. Which clue best distinguishes this issue from similar incidents?

  1. The plan includes hundreds of thousands of lookups for columns absent from the nonclustered index.
  2. The captured plan was compiled with a tiny tenant ID and scans badly when reused for a huge tenant.
  3. Performance traces show no index scan regression but rebuild operations consume hours of I/O.
  4. Resource charts show high CPU while the team has no ranked list of query costs.

Explanation

Correct Answer

B:

The captured plan was compiled with a tiny tenant ID and scans badly when reused for a huge tenant. Cardinality skew can make a single cached plan suboptimal across parameter values.

Incorrect Answers

A:

The plan includes hundreds of thousands of lookups for columns absent from the nonclustered index. Cardinality skew can make a single cached plan suboptimal across parameter values. The captured plan was compiled with a tiny tenant ID and scans badly when reused for a huge tenant.

C:

Performance traces show no index scan regression but rebuild operations consume hours of I/O. Cardinality skew can make a single cached plan suboptimal across parameter values. The captured plan was compiled with a tiny tenant ID and scans badly when reused for a huge tenant.

D:

Resource charts show high CPU while the team has no ranked list of query costs. Cardinality skew can make a single cached plan suboptimal across parameter values. The captured plan was compiled with a tiny tenant ID and scans badly when reused for a huge tenant.

Question 571

A regulated supplier wants independently verifiable evidence that recorded transaction adjustments have not been silently altered after approval. What is the underlying reason for the corrective step?

  1. A graph model can express connected entities but it is not automatically better for simple relational lookups.
  2. Semi-structured JSON still needs deliberate schema validation and indexing for predictable filtering.
  3. Ledger features provide tamper-evidence verification but do not replace authorization or backups.
  4. Sequences generate values independently of row identity and do not guarantee gapless numbering.

Explanation

Correct Answer

C:

Ledger features provide tamper-evidence verification but do not replace authorization or backups.

Incorrect Answers

A:

A graph model can express connected entities but it is not automatically better for simple relational lookups. Ledger features provide tamper-evidence verification but do not replace authorization or backups.

B:

Semi-structured JSON still needs deliberate schema validation and indexing for predictable filtering. Ledger features provide tamper-evidence verification but do not replace authorization or backups.

D:

Sequences generate values independently of row identity and do not guarantee gapless numbering. Ledger features provide tamper-evidence verification but do not replace authorization or backups.

Question 572

A customer portal sends retrieved SQL policy snippets to a model but discards source IDs, making claims impossible to audit. Which concept should guide the database developer in this situation?

  1. A RAG solution must acknowledge unsupported questions rather than invent source authority.
  2. Grounded answers require verifiable provenance, not merely retrieval in the background.
  3. A structured interchange format can make SQL-to-model data safer and less ambiguous than string concatenation.
  4. External REST calls from SQL must minimize disclosed data and enforce endpoint security.

Explanation

Correct Answer

B:

Grounded answers require verifiable provenance, not merely retrieval in the background.

Incorrect Answers

A:

A RAG solution must acknowledge unsupported questions rather than invent source authority. Grounded answers require verifiable provenance, not merely retrieval in the background.

C:

A structured interchange format can make SQL-to-model data safer and less ambiguous than string concatenation. Grounded answers require verifiable provenance, not merely retrieval in the background.

D:

External REST calls from SQL must minimize disclosed data and enforce endpoint security. Grounded answers require verifiable provenance, not merely retrieval in the background.

Question 573

A global incident system includes support notes in six languages and its English-only embedding model retrieves poor results for non-English tickets. Which TWO options together specify a corrective step and a meaningful test?

  1. Evaluate a CDC-based incremental vector update process with checkpoints and delete propagation.
  2. A held-out multilingual search set meets retrieval quality targets across required languages.
  3. Database transactions remain available during a model outage and embeddings catch up after recovery.
  4. Use an event-driven Azure Function SQL trigger pattern with durable retries and source version checks.
  5. Evaluate multilingual embedding models against labeled cross-language retrieval tests and relevant privacy constraints.

Explanation

Correct Answers

B:

A held-out multilingual search set meets retrieval quality targets across required languages. Evaluate multilingual embedding models against labeled cross-language retrieval tests and relevant privacy constraints.

E:

Evaluate multilingual embedding models against labeled cross-language retrieval tests and relevant privacy constraints.

Incorrect Answers

A:

Evaluate a CDC-based incremental vector update process with checkpoints and delete propagation. Evaluate multilingual embedding models against labeled cross-language retrieval tests and relevant privacy constraints.

C:

Database transactions remain available during a model outage and embeddings catch up after recovery. Evaluate multilingual embedding models against labeled cross-language retrieval tests and relevant privacy constraints. A held-out multilingual search set meets retrieval quality targets across required languages.

D:

Use an event-driven Azure Function SQL trigger pattern with durable retries and source version checks. Evaluate multilingual embedding models against labeled cross-language retrieval tests and relevant privacy constraints.

Question 574

An incentive report ranks top sellers by revenue but ties must follow a documented business tie-break rather than unpredictable row ordering. What should the database developer verify to confirm the suspected cause?

  1. The report shows a price amount from a different region than the chosen maximum date.
  2. Current substring slicing returns invalid tokens when the prefix appears in narrative words.
  3. The current scoring treats harmless name variations and unrelated records as equally likely duplicates.
  4. Two tied sellers swap order across executions despite identical source revenue.

Explanation

Correct Answer

D:

Two tied sellers swap order across executions despite identical source revenue. Ranking semantics must reflect whether ties share positions or require deterministic ordered rows.

Incorrect Answers

A:

The report shows a price amount from a different region than the chosen maximum date. Ranking semantics must reflect whether ties share positions or require deterministic ordered rows. Two tied sellers swap order across executions despite identical source revenue.

B:

Current substring slicing returns invalid tokens when the prefix appears in narrative words. Ranking semantics must reflect whether ties share positions or require deterministic ordered rows. Two tied sellers swap order across executions despite identical source revenue.

C:

The current scoring treats harmless name variations and unrelated records as equally likely duplicates. Ranking semantics must reflect whether ties share positions or require deterministic ordered rows. Two tied sellers swap order across executions despite identical source revenue.

Question 575

A subscription report needs the difference between current and prior month revenue per product without a complex self-join. Which corrective action best fits the observed symptoms?

  1. Evaluate REGEXP_REPLACE with a narrowly tested account-token pattern and independent sensitive-data detection.
  2. Benchmark JARO_WINKLER_DISTANCE or related supported measures against labeled name pairs.
  3. Use LAG partitioned by product and ordered by report month, then check handling for missing months.
  4. Use an appropriate window rank and secondary order according to the published tie rule.

Explanation

Correct Answer

C:

Use LAG partitioned by product and ordered by report month, then check handling for missing months. Window offset functions simplify ordered comparisons but do not create missing time periods.

Incorrect Answers

A:

Evaluate REGEXP_REPLACE with a narrowly tested account-token pattern and independent sensitive-data detection. Window offset functions simplify ordered comparisons but do not create missing time periods. Use LAG partitioned by product and ordered by report month, then check handling for missing months.

B:

Benchmark JARO_WINKLER_DISTANCE or related supported measures against labeled name pairs. Window offset functions simplify ordered comparisons but do not create missing time periods. Use LAG partitioned by product and ordered by report month, then check handling for missing months.

D:

Use an appropriate window rank and secondary order according to the published tie rule. Window offset functions simplify ordered comparisons but do not create missing time periods. Use LAG partitioned by product and ordered by report month, then check handling for missing months.

Question 576

An internal workflow needs a controlled request to invoke a narrowly scoped SQL procedure but developers expose ad hoc SQL execution through REST. Which clue best distinguishes this issue from similar incidents?

  1. A nested query fails to resolve variants even though the SQL foreign key relationships exist.
  2. Two updates between polls collapse into one current row and intermediate changes disappear.
  3. The production service starts with a default configuration missing the intended entity permissions.
  4. The generic endpoint accepts arbitrary SQL text from the client instead of a defined business operation.

Explanation

Correct Answer

D:

The generic endpoint accepts arbitrary SQL text from the client instead of a defined business operation. An API contract should expose bounded domain operations rather than unrestricted SQL execution.

Incorrect Answers

A:

A nested query fails to resolve variants even though the SQL foreign key relationships exist. An API contract should expose bounded domain operations rather than unrestricted SQL execution. The generic endpoint accepts arbitrary SQL text from the client instead of a defined business operation.

B:

Two updates between polls collapse into one current row and intermediate changes disappear. An API contract should expose bounded domain operations rather than unrestricted SQL execution. The generic endpoint accepts arbitrary SQL text from the client instead of a defined business operation.

C:

The production service starts with a default configuration missing the intended entity permissions. An API contract should expose bounded domain operations rather than unrestricted SQL execution. The generic endpoint accepts arbitrary SQL text from the client instead of a defined business operation.

Question 577

A developer commits an Azure SQL connection string with a password into a pipeline YAML and then deletes the line after review. Which test result would demonstrate the fix worked?

  1. The drift is reviewed, either incorporated or reverted, and deployment reports no unexplained difference.
  2. Old credentials are revoked, logs are controlled, and the pipeline authenticates without committed passwords.
  3. The pipeline rejects the faulty change and accepts the corrected procedure with verified results.
  4. CI produces a valid model and catches an intentional missing-object reference.

Explanation

Correct Answer

B:

Old credentials are revoked, logs are controlled, and the pipeline authenticates without committed passwords. The operating rule is: Deleting a committed secret does not remove the exposure or replace credential rotation.

Incorrect Answers

A:

The drift is reviewed, either incorporated or reverted, and deployment reports no unexplained difference. The operating rule is: Deleting a committed secret does not remove the exposure or replace credential rotation. Old credentials are revoked, logs are controlled, and the pipeline authenticates without committed passwords.

C:

The pipeline rejects the faulty change and accepts the corrected procedure with verified results. The operating rule is: Deleting a committed secret does not remove the exposure or replace credential rotation. Old credentials are revoked, logs are controlled, and the pipeline authenticates without committed passwords.

D:

CI produces a valid model and catches an intentional missing-object reference. The operating rule is: Deleting a committed secret does not remove the exposure or replace credential rotation. Old credentials are revoked, logs are controlled, and the pipeline authenticates without committed passwords.

Question 578

A T-SQL workflow builds a prompt for an external language model but includes every row and confidential identifier regardless of question scope. Choose TWO: one correct response and one verification of success.

  1. Test queries without evidence return a clearly marked gap and a human review route.
  2. The request sends only approved fields and the model receives sufficient task context.
  3. Include controlled document identifiers, version metadata, and explicit citation requirements in the RAG prompt.
  4. Use sp_invoke_external_rest_endpoint only with an approved network and identity configuration and a minimal validated prompt payload.
  5. Parse and validate the returned model response envelope and expected JSON fields before storing results.

Explanation

Correct Answers

B:

The request sends only approved fields and the model receives sufficient task context. Use sp_invoke_external_rest_endpoint only with an approved network and identity configuration and a minimal validated prompt payload.

D:

Use sp_invoke_external_rest_endpoint only with an approved network and identity configuration and a minimal validated prompt payload.

Incorrect Answers

A:

Test queries without evidence return a clearly marked gap and a human review route. Use sp_invoke_external_rest_endpoint only with an approved network and identity configuration and a minimal validated prompt payload.

C:

Include controlled document identifiers, version metadata, and explicit citation requirements in the RAG prompt. Use sp_invoke_external_rest_endpoint only with an approved network and identity configuration and a minimal validated prompt payload. The request sends only approved fields and the model receives sufficient task context.

E:

Parse and validate the returned model response envelope and expected JSON fields before storing results. Use sp_invoke_external_rest_endpoint only with an approved network and identity configuration and a minimal validated prompt payload. The request sends only approved fields and the model receives sufficient task context.

Question 579

Two feature branches independently modify the same customer table and the merged migration accidentally drops a required column. What should the database developer verify to confirm the suspected cause?

  1. The same stored credential can alter every production database regardless of the pipeline target.
  2. The merge compiles each branch separately but produces conflicting definitions in the combined model.
  3. A staged test user can query sensitive salary columns without being assigned an appropriate role.
  4. Pipeline logs identify the wrong connection target for a reviewed release.

Explanation

Correct Answer

B:

The merge compiles each branch separately but produces conflicting definitions in the combined model. Database merge conflicts can affect schema semantics beyond textual file overlap.

Incorrect Answers

A:

The same stored credential can alter every production database regardless of the pipeline target. Database merge conflicts can affect schema semantics beyond textual file overlap. The merge compiles each branch separately but produces conflicting definitions in the combined model.

C:

A staged test user can query sensitive salary columns without being assigned an appropriate role. Database merge conflicts can affect schema semantics beyond textual file overlap. The merge compiles each branch separately but produces conflicting definitions in the combined model.

D:

Pipeline logs identify the wrong connection target for a reviewed release. Database merge conflicts can affect schema semantics beyond textual file overlap. The merge compiles each branch separately but produces conflicting definitions in the combined model.

Question 580

A deployment script hardcodes a development database endpoint and CI unexpectedly connects to test during a production release. Which observation is the strongest evidence for the reported problem?

  1. Predeployment comparison shows the live procedure body differs from the reviewed project model.
  2. Pipeline logs identify the wrong connection target for a reviewed release.
  3. Repository history and build logs still contain the old secret even after the most recent file changes.
  4. The same stored credential can alter every production database regardless of the pipeline target.

Explanation

Correct Answer

B:

Pipeline logs identify the wrong connection target for a reviewed release. Deployment configuration and identity should jointly prevent cross-environment mistakes.

Incorrect Answers

A:

Predeployment comparison shows the live procedure body differs from the reviewed project model. Deployment configuration and identity should jointly prevent cross-environment mistakes. Pipeline logs identify the wrong connection target for a reviewed release.

C:

Repository history and build logs still contain the old secret even after the most recent file changes. Deployment configuration and identity should jointly prevent cross-environment mistakes. Pipeline logs identify the wrong connection target for a reviewed release.

D:

The same stored credential can alter every production database regardless of the pipeline target. Deployment configuration and identity should jointly prevent cross-environment mistakes. Pipeline logs identify the wrong connection target for a reviewed release.

Developing AI-Enabled Database Solutions (DP-800) Practice Tests