McAfee Secure

Exam Code: DP-800

Exam Name: Developing AI-Enabled Database Solutions

Certification Provider: Microsoft

Corresponding Certification: Microsoft Certified: SQL AI Developer Associate

Microsoft DP-800 Questions & Answers

Study with Up-To-Date REAL Exam Questions and Answers from the ACTUAL Test

101 Questions & Answers with Testing Engine

"Developing AI-Enabled Database Solutions Exam", also known as DP-800 exam, is a Microsoft certification exam.

Pass your tests with the always up-to-date DP-800 Exam Engine. Your DP-800 training materials keep you at the head of the pack!

Satisfaction Guaranteed

Money Back Guarantee

Testking provides hassle-free money back guarantee with our products. That is because we have 100% trust in the abilities of our professional and experience product team, and our record is a proof of that.

99.6% PASS RATE
Was: $137.49
Now: $124.99

Microsoft DP-800 Practice Test 83

Question 820

A reporting transaction under READ COMMITTED sees changing totals between consecutive statements. What is the MOST appropriate immediate corrective approach?

  1. Require an expected rowversion predicate and verify rows affected before confirming success.
  2. Identify the blocking head using supported DMV and wait analysis before scaling.
  3. Use relevant DMVs and Query Store runtime statistics to isolate resource-intensive statements.
  4. Evaluate SNAPSHOT isolation or appropriate transaction design for repeatable point-in-time reads.

Explanation

Correct Answer

D:

Evaluate SNAPSHOT isolation or appropriate transaction design for repeatable point-in-time reads. This targets the primary issue because isolation choices define which committed versions are visible across statements. Verification: Repeated statements in one snapshot transaction give a consistent view. Prevent recurrence: Monitor version store and update conflicts.

Incorrect Answers

A:

Optimistic concurrency only works if conflicts are detected, not silently overwritten.

B:

Lock waits are not resolved merely by allocating faster compute.

C:

Instance metrics alone do not identify which cached statements consume resources.

Question 821

An analyst wants to use ordinary LIKE predicates against randomized Always Encrypted ciphertext. What observed evidence would BEST confirm successful remediation?

  1. Supported equality or enclave test cases behave as designed without broad plaintext exposure.
  2. Authorized clients read/write after rotation and revoked keys no longer work.
  3. A spoofed request cannot change access to another tenant's rows.
  4. An export returns only rows permitted to its assigned role.

Explanation

Correct Answer

A:

Supported equality or enclave test cases behave as designed without broad plaintext exposure. This is the decisive observable evidence after implementing redesign the query or use an explicitly supported encryption/enclave pattern with documented tradeoffs. It tests the underlying requirement: Randomized encryption does not support general plaintext pattern matching inside the engine.

Incorrect Answers

B:

Authorized clients read/write after rotation and revoked keys no longer work. However, that result would not establish the required outcome: supported equality or enclave test cases behave as designed without broad plaintext exposure.

C:

A spoofed request cannot change access to another tenant's rows. However, that result would not establish the required outcome: supported equality or enclave test cases behave as designed without broad plaintext exposure.

D:

An export returns only rows permitted to its assigned role. However, that result would not establish the required outcome: supported equality or enclave test cases behave as designed without broad plaintext exposure.

Question 822

A SQL service's identity can call a model but lacks permission to read a required secure search index. What is the MOST relevant technical distinction in this situation?

  1. Embedding static credentials in SQL code can expose model access and billing.
  2. Identity permission at one Azure service does not imply access to adjacent resources.
  3. Untrusted headers cannot be treated as authenticated authorization claims.
  4. Passwordless CI reduces secret leakage and rotation burden when configured with scoped trust.

Explanation

Correct Answer

B:

Identity permission at one Azure service does not imply access to adjacent resources. This explains why the corrective action is to configure distinct least-privilege grants on model and search resources. Confirm the diagnosis when model calls and retrieval calls each succeed only with required scopes.

Incorrect Answers

A:

Embedding static credentials in SQL code can expose model access and billing. The actual technical constraint is that identity permission at one Azure service does not imply access to adjacent resources.

C:

Untrusted headers cannot be treated as authenticated authorization claims. The actual technical constraint is that identity permission at one Azure service does not imply access to adjacent resources.

D:

Passwordless CI reduces secret leakage and rotation burden when configured with scoped trust. The actual technical constraint is that identity permission at one Azure service does not imply access to adjacent resources.

Question 823

A recursive organization-chart query runs indefinitely after cyclic parent links are introduced. Which ongoing control is MOST relevant after the issue is fixed?

  1. Set sensible MAXRECURSION and detect cycles.
  2. Limit traversal depth and track source/target referential rules.
  3. Implement graph integrity checks on writes.
  4. Keep review logs and a reversible merge mechanism.

Explanation

Correct Answer

C:

Implement graph integrity checks on writes. This guards against recurrence because a cycle can repeatedly regenerate the same nodes in a recursive CTE. The direct remediation is: Validate acyclic parent relationships and bound recursion depth.

Incorrect Answers

A:

Set sensible MAXRECURSION and detect cycles. The current task instead requires this action: implement graph integrity checks on writes.

B:

Limit traversal depth and track source/target referential rules. The current task instead requires this action: implement graph integrity checks on writes.

D:

Keep review logs and a reversible merge mechanism. The current task instead requires this action: implement graph integrity checks on writes.

Question 824

A dev wants to enable all query optimization flags globally because one report is slow. What is the MOST appropriate immediate corrective approach?

  1. Measure workload resource patterns and choose an appropriate compute/service tier or scaling policy.
  2. Investigate parameter sensitivity, statistics and alternative plan strategies.
  3. Use Query Performance Insight and supported Azure/SQL monitoring evidence.
  4. Use targeted measured configuration changes and a rollback plan rather than blindly enabling trace flags.

Explanation

Correct Answer

D:

Use targeted measured configuration changes and a rollback plan rather than blindly enabling trace flags. This targets the primary issue because database-level settings can affect many workloads and engine versions differently. Verification: The report improves without regression in representative unrelated queries. Prevent recurrence: Preserve baseline settings and test in staging.

Incorrect Answers

A:

Measuring workload resource patterns and choose an appropriate compute/service tier or scaling policy would not correct a dev wants to enable all query optimization flags globally because one report is slow. Use targeted measured configuration changes and a rollback plan rather than blindly enabling trace flags.

B:

A poor cardinality estimate can induce unsuitable joins and memory grants.

C:

Platform-native query insight tools expose historical workload statistics for managed instances.

Question 825

A query uses vectors with unequal magnitudes and wants unit-length normalized comparisons. What observed evidence would BEST confirm successful remediation?

  1. Normalized test vectors have unit norm within floating tolerance.
  2. A known test vector reports the correct dimension property.
  3. Cross-tenant queries expose no IDs or snippets from denied records.
  4. Results contain one entry per source passage in a stable ranking.

Explanation

Correct Answer

A:

Normalized test vectors have unit norm within floating tolerance. This is the decisive observable evidence after implementing use VECTOR_NORMALIZE where supported before the intended comparison workflow. It tests the underlying requirement: Normalizing vector length separates direction from magnitude for suitable similarity metrics.

Incorrect Answers

B:

A known test vector reports the correct dimension property. However, that result would not establish the required outcome: normalized test vectors have unit norm within floating tolerance.

C:

Cross-tenant queries expose no IDs or snippets from denied records. However, that result would not establish the required outcome: normalized test vectors have unit norm within floating tolerance.

D:

Results contain one entry per source passage in a stable ranking. However, that result would not establish the required outcome: normalized test vectors have unit norm within floating tolerance.

Question 826

A salesperson needs to compare each month's sales with the prior month's result. Which technical reason BEST explains the proposed correction?

  1. Window ranking order is defined by its OVER clause, not the outer result sort.
  2. LAG references a preceding ordered row without self-joining every record.
  3. JSON_VALUE extracts a scalar path, while OPENJSON can shred collections.
  4. Graph relationships support path queries but still need guarded edge permissions.

Explanation

Correct Answer

B:

LAG references a preceding ordered row without self-joining every record. This explains why the corrective action is to use LAG(SalesAmount) OVER(PARTITION BY Region ORDER BY SalesMonth). Confirm the diagnosis when month-over-month differences match a reference table.

Incorrect Answers

A:

Window ranking order is defined by its OVER clause, not the outer result sort. The actual technical constraint is that LAG references a preceding ordered row without self-joining every record.

C:

JSON_VALUE extracts a scalar path, while OPENJSON can shred collections. The actual technical constraint is that LAG references a preceding ordered row without self-joining every record.

D:

Graph relationships support path queries but still need guarded edge permissions. The actual technical constraint is that LAG references a preceding ordered row without self-joining every record.

Question 827

A service trusts an unsigned client-supplied user role in a REST header. What maintenance or governance control BEST reduces recurrence of this problem?

  1. Combine DAB permissions with SQL RLS where applicable.
  2. Do not use masking as the sole defense for highly sensitive fields.
  3. Rotate signing keys and validate audiences and issuers.
  4. Review grants after schema migrations.

Explanation

Correct Answer

C:

Rotate signing keys and validate audiences and issuers. This guards against recurrence because untrusted headers cannot be treated as authenticated authorization claims. The direct remediation is: Validate provider-issued identity tokens and map claims to authorized application roles.

Incorrect Answers

A:

Combine DAB permissions with SQL RLS where applicable. The current task instead requires this action: rotate signing keys and validate audiences and issuers.

B:

Do not use masking as the sole defense for highly sensitive fields.

D:

Reviewing grants after schema migrations addresses a different database behavior; the relevant distinction is that rotate signing keys and validate audiences and issuers.

Question 828

A developer tries to use CONTAINS on a text column without a full-text index. Which implementation decision BEST addresses this requirement?

  1. Use a suitable CONTAINS phrase predicate for literal phrase intent.
  2. Use VECTOR_NORMALIZE where supported before the intended comparison workflow.
  3. Compute compatible query vector and use supported VECTOR_SEARCH or exact top-k distance query with authorization filters.
  4. Create a supported full-text index and required unique key on the searchable table.

Explanation

Correct Answer

D:

Create a supported full-text index and required unique key on the searchable table. This targets the primary issue because full-text predicates require a configured index and supported column types. Verification: A controlled CONTAINS query succeeds after population completes. Prevent recurrence: Inspect index freshness and population failures.

Incorrect Answers

A:

FREETEXT expands linguistic meaning and is not equivalent to exact phrase matching.

B:

Normalizing vector length separates direction from magnitude for suitable similarity metrics.

C:

Vector search ranks stored embeddings by distance but must honor source permissions.

Question 829

An auditor sees database logs kept on the same instance a compromised admin can delete. Which measurable outcome BEST verifies that the requirement is satisfied?

  1. A simulated service administrator cannot erase the retained central record.
  2. A test UPDATE appears in protected audit storage with correct identity.
  3. Supported equality or enclave test cases behave as designed without broad plaintext exposure.
  4. Authorized clients read/write after rotation and revoked keys no longer work.

Explanation

Correct Answer

A:

A simulated service administrator cannot erase the retained central record. This is the decisive observable evidence after implementing send audit output to appropriately protected centralized storage with restricted permissions. It tests the underlying requirement: Tamper-resistant off-host retention is stronger evidence than only local logs.

Incorrect Answers

B:

A test UPDATE appears in protected audit storage with correct identity. However, that result would not establish the required outcome: a simulated service administrator cannot erase the retained central record.

C:

Supported equality or enclave test cases behave as designed without broad plaintext exposure. However, that result would not establish the required outcome: a simulated service administrator cannot erase the retained central record.

D:

Authorized clients read/write after rotation and revoked keys no longer work. However, that result would not establish the required outcome: a simulated service administrator cannot erase the retained central record.

Question 830

A finance application needs deterministic machine-readable response fields from model inference. Which technical reason BEST explains the proposed correction?

  1. Model size increases cost and latency without guaranteeing proportional quality gains.
  2. Unconstrained prose is fragile to parse and can omit required keys.
  3. Deterministic chunk identity reduces unnecessary inference and preserves provenance.
  4. Frequent numeric fields add update cost and can weaken semantic relevance.

Explanation

Correct Answer

B:

Unconstrained prose is fragile to parse and can omit required keys. This explains why the corrective action is to prefer a model and API mode supporting validated structured output and enforce schema checks. Confirm the diagnosis when sample calls produce valid JSON schema instances or explicit errors.

Incorrect Answers

A:

Modeling size increases cost and latency without guaranteeing proportional quality gains concerns a different mechanism; the relevant technical point is that unconstrained prose is fragile to parse and can omit required keys.

C:

Deterministic chunk identity reduces unnecessary inference and preserves provenance. The actual technical constraint is that unconstrained prose is fragile to parse and can omit required keys.

D:

Frequent numeric fields add update cost and can weaken semantic relevance. The actual technical constraint is that unconstrained prose is fragile to parse and can omit required keys.

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