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 17

Question 160

A stored procedure passes DECIMAL prices as unquoted strings with locale commas and model summaries misstate values. What maintenance or governance control BEST reduces recurrence of this problem?

  1. Enforce outbound allowlists and unit tests for fields.
  2. Keep business arithmetic outside uncontrolled model generation.
  3. Log safe diagnostics and maintain circuit-breaker policy.
  4. Compare model output to authoritative SQL totals.

Explanation

Correct Answer

D:

Compare model output to authoritative SQL totals. This guards against recurrence because JSON encoding preserves numeric representation better than locale-dependent concatenation. The direct remediation is: Serialize numeric SQL types as valid JSON numbers with explicit unit metadata.

Incorrect Answers

A:

Enforce outbound allowlists and unit tests for fields. The current task instead requires this action: compare model output to authoritative SQL totals.

B:

Keep business arithmetic outside uncontrolled model generation. The current task instead requires this action: compare model output to authoritative SQL totals.

C:

Log safe diagnostics and maintain circuit-breaker policy. The current task instead requires this action: compare model output to authoritative SQL totals.

Question 161

One document ranks first in both lexical and vector results while another appears only ninth in one. What should the SQL developer do FIRST to address the specific issue?

  1. Use RRF to reward documents ranking highly across multiple candidate lists.
  2. Compute ranks within each result list before applying reciprocal-rank contributions.
  3. Generate query embeddings with the same approved model and dimension as the column.
  4. Use indexed lexical/full-text or exact filter matching appropriate to the identifier.

Explanation

Correct Answer

A:

Use RRF to reward documents ranking highly across multiple candidate lists. This targets the primary issue because RRF sums reciprocal functions of positions across lists. Verification: The joint top-ranked item receives the expected combined score. Prevent recurrence: Avoid counting duplicates in one list twice.

Incorrect Answers

B:

RRF operates on list positions and is distinct from raw score interpolation.

C:

Generate query embeddings with the same approved model and dimension as the column. The relevant technical condition instead is that use RRF to reward documents ranking highly across multiple candidate lists.

D:

Embedding similarity can confuse near-looking product codes that must match exactly.

Question 162

An application needs encrypted values at rest but server-side procedures must decrypt a small field. Which acceptance check is MOST directly related to this scenario?

  1. A privileged SQL-only session cannot read plaintext while an approved client decrypts.
  2. Approved procedures decrypt the target values and unauthorized roles cannot open keys.
  3. A spoofed request cannot change access to another tenant's rows.
  4. Anonymous calls fail while authorized SQL workloads succeed.

Explanation

Correct Answer

B:

Approved procedures decrypt the target values and unauthorized roles cannot open keys. This is the decisive observable evidence after implementing evaluate SQL column-level encryption with explicit key and certificate controls. It tests the underlying requirement: Server-side encryption can decrypt inside SQL with authorized key access, unlike client-controlled Always Encrypted.

Incorrect Answers

A:

A privileged SQL-only session cannot read plaintext while an approved client decrypts. However, that result would not establish the required outcome: approved procedures decrypt the target values and unauthorized roles cannot open keys.

C:

A spoofed request cannot change access to another tenant's rows. However, that result would not establish the required outcome: approved procedures decrypt the target values and unauthorized roles cannot open keys.

D:

Anonymous calls fail while authorized SQL workloads succeed. However, that result would not establish the required outcome: approved procedures decrypt the target values and unauthorized roles cannot open keys.

Question 163

An RLS policy uses SESSION_CONTEXT set directly from untrusted request text. Which technical property or constraint MOST directly matters here?

  1. Security enforcement must be tested with the exact executing principal.
  2. API generation does not automatically imply that exposed data is appropriately authorized.
  3. Session context is only safe as an authorization input when callers cannot impersonate other tenants.
  4. Server-side encryption can decrypt inside SQL with authorized key access, unlike client-controlled Always Encrypted.

Explanation

Correct Answer

C:

Session context is only safe as an authorization input when callers cannot impersonate other tenants. This explains why the corrective action is to derive tenant identity from verified authentication and tightly controlled session setup. Confirm the diagnosis when a spoofed request cannot change access to another tenant's rows.

Incorrect Answers

A:

Security enforcement must be tested with the exact executing principal. The actual technical constraint is that session context is only safe as an authorization input when callers cannot impersonate other tenants.

B:

API generation does not automatically imply that exposed data is appropriately authorized. The actual technical constraint is that session context is only safe as an authorization input when callers cannot impersonate other tenants.

D:

Server-side encryption can decrypt inside SQL with authorized key access, unlike client-controlled Always Encrypted. The actual technical constraint is that session context is only safe as an authorization input when callers cannot impersonate other tenants.

Question 164

A deployment blindly INSERTs static country rows on every run and fails with duplicates. Which preventive practice BEST addresses the underlying risk over time?

  1. Rotate exposed secrets and enable precommit scanning.
  2. Run deployment reports and a verified rollback strategy.
  3. Keep isolated test credentials and avoid production data.
  4. Avoid overwriting locally maintained fields unexpectedly.

Explanation

Correct Answer

D:

Avoid overwriting locally maintained fields unexpectedly. This guards against recurrence because repeatable deployments must not assume reference rows are absent. The direct remediation is: Use idempotent keyed upserts or appropriate repeat-safe seed operations.

Incorrect Answers

A:

Rotate exposed secrets and enable precommit scanning. The current task instead requires this action: avoid overwriting locally maintained fields unexpectedly.

B:

Run deployment reports and a verified rollback strategy. The current task instead requires this action: avoid overwriting locally maintained fields unexpectedly.

C:

Keep isolated test credentials and avoid production data. The current task instead requires this action: avoid overwriting locally maintained fields unexpectedly.

Question 165

An application expects a 64-bit vendor identifier but its SQL key column is INT and may overflow. What should the SQL developer do FIRST to address the specific issue?

  1. Select BIGINT after modeling value range, foreign keys and migration effects.
  2. Replace FLOAT money arithmetic with appropriately sized DECIMAL precision and scale.
  3. Evaluate memory-optimized tables where the specific SQL engine and workload support them.
  4. Choose a system-versioned temporal table with properly managed history and retention.

Explanation

Correct Answer

A:

Select BIGINT after modeling value range, foreign keys and migration effects. This targets the primary issue because an INT cannot represent IDs above its signed range. Verification: High-ID inserts and all dependent joins work in migration tests. Prevent recurrence: Validate client parameter types and index-size effects before deployment.

Incorrect Answers

B:

Binary floating-point approximations cannot represent every decimal currency value exactly.

C:

Memory-optimized OLTP can reduce contention but adds feature and durability constraints.

D:

Temporal history captures prior row versions when system versioning is enabled.

Question 166

An order detail table is queried by CustomerId and CreatedAt and performs repeated full scans. Which acceptance check is MOST directly related to this scenario?

  1. Reconciled payment totals match control calculations to the cent.
  2. The representative query reads fewer pages and retains correct ordered rows.
  3. Sample records validate JSON paths while required keys remain enforced.
  4. A benchmark improves queue throughput while meeting durability requirements.

Explanation

Correct Answer

B:

The representative query reads fewer pages and retains correct ordered rows. This is the decisive observable evidence after implementing design a useful composite rowstore index with leading keys matching the filters and inspect the execution plan. It tests the underlying requirement: A missing selective access path forces large table or clustered-index scans.

Incorrect Answers

A:

Reconciled payment totals match control calculations to the cent. However, that result would not establish the required outcome: the representative query reads fewer pages and retains correct ordered rows.

C:

Sample records validate JSON paths while required keys remain enforced. However, that result would not establish the required outcome: the representative query reads fewer pages and retains correct ordered rows.

D:

A benchmark improves queue throughput while meeting durability requirements. However, that result would not establish the required outcome: the representative query reads fewer pages and retains correct ordered rows.

Question 167

A user asks for 'how to prevent late shipment penalties' but document text says 'service-level liquidated damages'. Which technical property or constraint MOST directly matters here?

  1. Embedding similarity can confuse near-looking product codes that must match exactly.
  2. RRF sums reciprocal functions of positions across lists.
  3. Embedding search can retrieve related ideas even without matching literal tokens.
  4. Native vector types validate dimensions and avoid repeated parsing of text arrays.

Explanation

Correct Answer

C:

Embedding search can retrieve related ideas even without matching literal tokens. This explains why the corrective action is to use semantic vector retrieval to recognize conceptual meaning beyond exact words. Confirm the diagnosis when the known penalty clause ranks in a held-out test query.

Incorrect Answers

A:

Embedding similarity can confuse near-looking product codes that must match exactly. The actual technical constraint is that embedding search can retrieve related ideas even without matching literal tokens.

B:

RRF sums reciprocal functions of positions across lists. The actual technical constraint is that embedding search can retrieve related ideas even without matching literal tokens.

D:

Native vector types validate dimensions and avoid repeated parsing of text arrays. The actual technical constraint is that embedding search can retrieve related ideas even without matching literal tokens.

Question 168

A Copilot MCP configuration points at a stale server with changed schema metadata. Following remediation, what safeguard BEST prevents a repeat of this specific failure?

  1. Maintain a controlled model catalog and audit changes.
  2. Protect branches with required security code review.
  3. Do not grant production DDL permissions directly to AI tools.
  4. Pin trusted endpoints and monitor certificate or token changes.

Explanation

Correct Answer

D:

Pin trusted endpoints and monitor certificate or token changes. This guards against recurrence because tool discovery and endpoint configuration determine accessible operations. The direct remediation is: Update and validate the MCP connection and available tool schema in the client.

Incorrect Answers

A:

Maintain a controlled model catalog and audit changes. The current task instead requires this action: pin trusted endpoints and monitor certificate or token changes.

B:

Protect branches with required security code review. The current task instead requires this action: pin trusted endpoints and monitor certificate or token changes.

C:

Do not grant production DDL permissions directly to AI tools.

Question 169

A product API caches inventory counts for an hour and customers see stale stock after orders. Which technical action BEST resolves the primary problem?

  1. Use an appropriate DAB cache TTL or disable cache for volatile inventory fields.
  2. Configure controlled pagination, sorting and maximum page size for the DAB entity.
  3. Configure supported network access, managed identity and database permissions.
  4. Parameterize the DAB data source using supported environment variables or secret integration.

Explanation

Correct Answer

A:

Use an appropriate DAB cache TTL or disable cache for volatile inventory fields. This targets the primary issue because response caching trades freshness for latency and may serve obsolete facts. Verification: After a stock change, clients see updated counts within the defined freshness SLA. Prevent recurrence: Document invalidation and avoid caching sensitive responses incorrectly.

Incorrect Answers

B:

Unbounded result sets waste bandwidth and can overload SQL and API memory.

C:

Publishing the API does not grant the host network or SQL access automatically.

D:

Configuration must resolve the correct database per environment without leaking secrets.

Question 170

Queries filter millions of product rows by a JSON property and repeatedly scan payloads. Which test result BEST demonstrates that the intended solution is working?

  1. A test payload creates the expected number of items with correct types.
  2. The execution plan uses the chosen path access strategy and returns matching products.
  3. A test reserves the requested count and avoids collisions.
  4. A maintenance rehearsal successfully switches a partition with required constraints.

Explanation

Correct Answer

B:

The execution plan uses the chosen path access strategy and returns matching products. This is the decisive observable evidence after implementing evaluate supported JSON index or indexed computed-column path for that SQL platform and version. It tests the underlying requirement: Unindexed JSON path predicates can incur expensive row-by-row parsing.

Incorrect Answers

A:

A test payload creates the expected number of items with correct types. However, that result would not establish the required outcome: the execution plan uses the chosen path access strategy and returns matching products.

C:

A test reserves the requested count and avoids collisions. However, that result would not establish the required outcome: the execution plan uses the chosen path access strategy and returns matching products.

D:

A maintenance rehearsal successfully switches a partition with required constraints. However, that result would not establish the required outcome: the execution plan uses the chosen path access strategy and returns matching products.

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