Microsoft DP-800 Practice Test 39 - 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 39

Question 380

A support platform stores optional case attributes in a JSON column and filters on a frequent nested risk-level property at scale. Choose TWO statements: a suitable intervention and evidence of its success.

  1. Design a governed temporal retention and deletion procedure across current and historical data.
  2. A representative graph traversal returns validated connections at acceptable cost.
  3. Design JSON access and indexing according to the engine's supported JSON features and test the predicate.
  4. Queries find the correct risk values with less CPU and no loss of document validity.
  5. Consider supported graph node and edge tables and evaluate query readability and performance against relational alternatives.

Explanation

Correct Answers

C:

Design JSON access and indexing according to the engine's supported JSON features and test the predicate.

D:

Queries find the correct risk values with less CPU and no loss of document validity. Design JSON access and indexing according to the engine's supported JSON features and test the predicate.

Incorrect Answers

A:

Design a governed temporal retention and deletion procedure across current and historical data. Design JSON access and indexing according to the engine's supported JSON features and test the predicate.

B:

A representative graph traversal returns validated connections at acceptable cost. Design JSON access and indexing according to the engine's supported JSON features and test the predicate. Queries find the correct risk values with less CPU and no loss of document validity.

E:

Consider supported graph node and edge tables and evaluate query readability and performance against relational alternatives. Design JSON access and indexing according to the engine's supported JSON features and test the predicate. Queries find the correct risk values with less CPU and no loss of document validity.

Question 381

A Data API builder endpoint accepts unvalidated client-side tenant parameters while backend queries use an all-tenant database identity. Which outcome shows the requirement is met without introducing a new risk?

  1. The application connects with its scoped identity while no reusable database password is stored in code.
  2. Modified client filters cannot expand the account's authorized row scope.
  3. Unapproved GraphQL fields remain unavailable while normal approved queries operate.
  4. Compromised low-privilege identities cannot read unrelated protected data in a controlled test.

Explanation

Correct Answer

B:

Modified client filters cannot expand the account's authorized row scope. The intended intervention is: Enforce authorized tenant scoping through backend policies and service identity rather than trust the caller's filter.

Incorrect Answers

A:

The application connects with its scoped identity while no reusable database password is stored in code. The intended intervention is: Enforce authorized tenant scoping through backend policies and service identity rather than trust the caller's filter. Modified client filters cannot expand the account's authorized row scope.

C:

Unapproved GraphQL fields remain unavailable while normal approved queries operate. The intended intervention is: Enforce authorized tenant scoping through backend policies and service identity rather than trust the caller's filter. Modified client filters cannot expand the account's authorized row scope.

D:

Compromised low-privilege identities cannot read unrelated protected data in a controlled test. The intended intervention is: Enforce authorized tenant scoping through backend policies and service identity rather than trust the caller's filter. Modified client filters cannot expand the account's authorized row scope.

Question 382

A mobile inventory API requires a JSON object per warehouse containing an array of item summaries, and string concatenation produces malformed output for quoted names. What is the underlying reason for the corrective step?

  1. Native JSON functions help create valid structured output instead of fragile manual string concatenation.
  2. Ordered window aggregates express running calculations without repeated per-row rescans.
  3. Window functions can rank related rows while retaining access to complete row values.
  4. Ranking semantics must reflect whether ties share positions or require deterministic ordered rows.

Explanation

Correct Answer

A:

Native JSON functions help create valid structured output instead of fragile manual string concatenation.

Incorrect Answers

B:

Ordered window aggregates express running calculations without repeated per-row rescans. Native JSON functions help create valid structured output instead of fragile manual string concatenation.

C:

Window functions can rank related rows while retaining access to complete row values. Native JSON functions help create valid structured output instead of fragile manual string concatenation.

D:

Ranking semantics must reflect whether ties share positions or require deterministic ordered rows. Native JSON functions help create valid structured output instead of fragile manual string concatenation.

Question 383

An Azure-hosted service connects to Azure SQL with a long-lived database password embedded in a configuration file and rotates credentials infrequently. Which concept should guide the database developer in this situation?

  1. Model endpoints are security-sensitive downstream services and credentials must be governed.
  2. Security auditing requires deliberate event selection, integrity, and lifecycle management.
  3. Encryption at rest does not replace database access controls for authenticated query operations.
  4. Passwordless managed identity can reduce secret distribution when authorization is configured correctly.

Explanation

Correct Answer

D:

Passwordless managed identity can reduce secret distribution when authorization is configured correctly.

Incorrect Answers

A:

Model endpoints are security-sensitive downstream services and credentials must be governed. Passwordless managed identity can reduce secret distribution when authorization is configured correctly.

B:

Security auditing requires deliberate event selection, integrity, and lifecycle management. Passwordless managed identity can reduce secret distribution when authorization is configured correctly.

C:

Encryption at rest does not replace database access controls for authenticated query operations. Passwordless managed identity can reduce secret distribution when authorization is configured correctly.

Question 384

Developers can push directly to the default branch and deploy payment-table changes without review or security testing. Which outcome shows the requirement is met without introducing a new risk?

  1. The merged project builds and a staged migration preserves all approved fields.
  2. A tested failing pull request is blocked and approved changes follow the defined merge path.
  3. The drift is reviewed, either incorporated or reverted, and deployment reports no unexplained difference.
  4. Both old and new clients function through the transition and rollback works in a rehearsed test.

Explanation

Correct Answer

B:

A tested failing pull request is blocked and approved changes follow the defined merge path. Developers can push directly to the default branch and deploy payment-table changes without review or security testing.

Incorrect Answers

A:

The merged project builds and a staged migration preserves all approved fields. Developers can push directly to the default branch and deploy payment-table changes without review or security testing. A tested failing pull request is blocked and approved changes follow the defined merge path.

C:

The drift is reviewed, either incorporated or reverted, and deployment reports no unexplained difference. Developers can push directly to the default branch and deploy payment-table changes without review or security testing. A tested failing pull request is blocked and approved changes follow the defined merge path.

D:

Both old and new clients function through the transition and rollback works in a rehearsed test. Developers can push directly to the default branch and deploy payment-table changes without review or security testing. A tested failing pull request is blocked and approved changes follow the defined merge path.

Question 385

A team puts a remote inference call inside a database trigger so every customer update must wait for an embedding service response. Which observation is the strongest evidence for the reported problem?

  1. The vector table retains deleted product entries and lacks vectors for recent updates.
  2. The same document produces different field names and invalid JSON across retries.
  3. An outage at the model endpoint causes ordinary customer updates to fail or stall.
  4. An edited policy record has a current text version but the vector represents an older summary.

Explanation

Correct Answer

C:

An outage at the model endpoint causes ordinary customer updates to fail or stall. Synchronous external calls inside write transactions can couple availability to a remote inference service.

Incorrect Answers

A:

The vector table retains deleted product entries and lacks vectors for recent updates. Synchronous external calls inside write transactions can couple availability to a remote inference service. An outage at the model endpoint causes ordinary customer updates to fail or stall.

B:

The same document produces different field names and invalid JSON across retries. Synchronous external calls inside write transactions can couple availability to a remote inference service. An outage at the model endpoint causes ordinary customer updates to fail or stall.

D:

An edited policy record has a current text version but the vector represents an older summary. Synchronous external calls inside write transactions can couple availability to a remote inference service. An outage at the model endpoint causes ordinary customer updates to fail or stall.

Question 386

A database team embeds unredacted personal case notes into a vector store and assumes the result cannot expose personal information because it is numeric. Which test result would demonstrate the fix worked?

  1. The function processes authorized events while negative access tests deny unrelated tables.
  2. Attempted destructive actions are blocked and permitted read tools function for authorized identities.
  3. Approved salary searches work through a supported secure design without weakening confidentiality.
  4. Only approved fields are embedded and unauthorized users cannot retrieve protected source or embedding records.

Explanation

Correct Answer

D:

Only approved fields are embedded and unauthorized users cannot retrieve protected source or embedding records. The operating rule is: Embeddings can retain sensitive information relationships and must follow the underlying data governance.

Incorrect Answers

A:

The function processes authorized events while negative access tests deny unrelated tables. The operating rule is: Embeddings can retain sensitive information relationships and must follow the underlying data governance. Only approved fields are embedded and unauthorized users cannot retrieve protected source or embedding records.

B:

Attempted destructive actions are blocked and permitted read tools function for authorized identities. The operating rule is: Embeddings can retain sensitive information relationships and must follow the underlying data governance. Only approved fields are embedded and unauthorized users cannot retrieve protected source or embedding records.

C:

Approved salary searches work through a supported secure design without weakening confidentiality. The operating rule is: Embeddings can retain sensitive information relationships and must follow the underlying data governance. Only approved fields are embedded and unauthorized users cannot retrieve protected source or embedding records.

Question 387

A pipeline uses a shared administrator password to publish SQL Database Project changes to multiple production databases. What should the database developer verify to confirm the suspected cause?

  1. A code review finds an untested branch that grants excessive discounts when a value is missing.
  2. The same stored credential can alter every production database regardless of the pipeline target.
  3. Predeployment comparison shows the live procedure body differs from the reviewed project model.
  4. Pipeline logs identify the wrong connection target for a reviewed release.

Explanation

Correct Answer

B:

The same stored credential can alter every production database regardless of the pipeline target. Automated deployment identities should be scoped to the resources and actions they operate.

Incorrect Answers

A:

A code review finds an untested branch that grants excessive discounts when a value is missing. Automated deployment identities should be scoped to the resources and actions they operate. The same stored credential can alter every production database regardless of the pipeline target.

C:

Predeployment comparison shows the live procedure body differs from the reviewed project model. Automated deployment identities should be scoped to the resources and actions they operate. The same stored credential can alter every production database regardless of the pipeline target.

D:

Pipeline logs identify the wrong connection target for a reviewed release. Automated deployment identities should be scoped to the resources and actions they operate. The same stored credential can alter every production database regardless of the pipeline target.

Question 388

A retrieved customer comment contains instructions telling the model to execute a database update and the application passes that text into the generation prompt. Select TWO: the best action and the outcome that confirms it worked.

  1. The model summarizes the comment without executing any unauthorized database change.
  2. A controlled request cites the active policy and does not require a model retraining cycle.
  3. Treat retrieved content as untrusted evidence and never grant it authority to execute SQL actions.
  4. Record authorized prompt/source version metadata, model identity, and approval outcome with appropriate privacy retention.
  5. Use sp_invoke_external_rest_endpoint only with an approved network and identity configuration and a minimal validated prompt payload.

Explanation

Correct Answers

A:

The model summarizes the comment without executing any unauthorized database change. Treat retrieved content as untrusted evidence and never grant it authority to execute SQL actions.

C:

Treat retrieved content as untrusted evidence and never grant it authority to execute SQL actions.

Incorrect Answers

B:

A controlled request cites the active policy and does not require a model retraining cycle. Treat retrieved content as untrusted evidence and never grant it authority to execute SQL actions. The model summarizes the comment without executing any unauthorized database change.

D:

Record authorized prompt/source version metadata, model identity, and approval outcome with appropriate privacy retention. Treat retrieved content as untrusted evidence and never grant it authority to execute SQL actions. The model summarizes the comment without executing any unauthorized database change.

E:

Use sp_invoke_external_rest_endpoint only with an approved network and identity configuration and a minimal validated prompt payload. Treat retrieved content as untrusted evidence and never grant it authority to execute SQL actions. The model summarizes the comment without executing any unauthorized database change.

Question 389

A SQL AI search service is deemed successful because every query returns some result, even when relevant records rarely appear in the first page. What should the database developer do FIRST within approved procedures?

  1. Build a labeled query set and track recall at k, precision, latency, and failure slices.
  2. Compare ANN and exact-nearest-neighbor methods using agreed recall and latency thresholds.
  3. Evaluate supported full-text indexing and query patterns with language-specific tokenization requirements.
  4. Review supported filter placement and retrieve enough authorized candidates before ranking as appropriate.

Explanation

Correct Answer

A:

Build a labeled query set and track recall at k, precision, latency, and failure slices. Search evaluation needs relevance metrics rather than simple nonempty result counts.

Incorrect Answers

B:

Compare ANN and exact-nearest-neighbor methods using agreed recall and latency thresholds. Search evaluation needs relevance metrics rather than simple nonempty result counts. Build a labeled query set and track recall at k, precision, latency, and failure slices.

C:

Evaluate supported full-text indexing and query patterns with language-specific tokenization requirements. Search evaluation needs relevance metrics rather than simple nonempty result counts. Build a labeled query set and track recall at k, precision, latency, and failure slices.

D:

Review supported filter placement and retrieve enough authorized candidates before ranking as appropriate. Search evaluation needs relevance metrics rather than simple nonempty result counts. Build a labeled query set and track recall at k, precision, latency, and failure slices.

Question 390

An API accepts arbitrary malformed JSON in a text field and downstream reporting intermittently fails during JSON parsing. Which technical principle explains the correct response?

  1. Columnstore health depends on ingestion and modification patterns as well as index creation.
  2. A graph model can express connected entities but it is not automatically better for simple relational lookups.
  3. Data shape and validity should be enforced as close to ingestion as practical.
  4. Ledger features provide tamper-evidence verification but do not replace authorization or backups.

Explanation

Correct Answer

C:

Data shape and validity should be enforced as close to ingestion as practical.

Incorrect Answers

A:

Columnstore health depends on ingestion and modification patterns as well as index creation. Data shape and validity should be enforced as close to ingestion as practical.

B:

A graph model can express connected entities but it is not automatically better for simple relational lookups. Data shape and validity should be enforced as close to ingestion as practical.

D:

Ledger features provide tamper-evidence verification but do not replace authorization or backups. Data shape and validity should be enforced as close to ingestion as practical.

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