Implementing Data Engineering Solutions Using Azure Databricks DP-750 Practice Test 4
Question 30
A Databricks engineer recommends: Split ingestion, validation and publishing into explicit orchestrated tasks with durable intermediate outputs. It responds to: A multi-step notebook workflow handles source extraction and final reporting in one cell sequence with no recoverable checkpoints. What underlying mechanism BEST explains why?
- Task-level repair reduces unnecessary recomputation while preserving dependency correctness.
- Job status must accurately reflect failed required operations for alerting and dependency logic.
- Retryable transient failures should be distinguished from permanent data or permission errors.
- Separating pipeline stages creates observable failure boundaries and supports targeted recovery.
Explanation
Correct Answer
D:The decisive mechanism is: Separating pipeline stages creates observable failure boundaries and supports targeted recovery. That explains the selected action: Split ingestion, validation and publishing into explicit orchestrated tasks with durable intermediate outputs. In the original scenario: A multi-step notebook workflow handles source extraction and final reporting in one cell sequence with no recoverable checkpoints. The specific unmanaged risk is: One late-stage error may force a costly replay of all preceding successful processing. Validate with: A simulated final-task failure can be repaired without rerunning already validated upstream ingestion.
Incorrect Answers
A:This principle concerns a separate data-engineering condition: A bulk repair executes all pipeline tasks again even when only the final reporting notebook failed. Its targeted response is: Separate tasks and use the supported repair-run workflow at the failed stage when dependencies remain valid. The current case is governed more directly by: Separating pipeline stages creates observable failure boundaries and supports targeted recovery.
B:This principle concerns a separate data-engineering condition: A workflow uses a notebook task that catches exceptions and returns success even when a write step fails. Its targeted response is: Propagate nonrecoverable exceptions to Lakeflow Jobs and attach actionable error metadata. The current case is governed more directly by: Separating pipeline stages creates observable failure boundaries and supports targeted recovery.
C:This principle concerns a separate data-engineering condition: A batch enrichment task downloads a remote file and sometimes receives transient HTTP 503 responses. Its targeted response is: Add bounded retries with exponential backoff and observable failure after the retry budget. The current case is governed more directly by: Separating pipeline stages creates observable failure boundaries and supports targeted recovery.
Question 31
For this Microsoft DP-750 case: A cross-engine consumer expects a Delta table but source is plain Iceberg and engineers assume identical mutation semantics. Which design or operational response BEST fits?
- Choose and document a supported format and catalog integration with explicit read/write compatibility tests.
- Enable supported liquid clustering with appropriate keys and run compatible optimization.
- Choose incremental append or merge based on immutable versus mutable source behavior.
- Select CTAS for an explicitly materialized snapshot or a scheduled pipeline for ongoing refresh.
Explanation
Correct Answer
A:The most appropriate action is: Choose and document a supported format and catalog integration with explicit read/write compatibility tests. It addresses the stated condition: A cross-engine consumer expects a Delta table but source is plain Iceberg and engineers assume identical mutation semantics. The technical basis is: Delta and Iceberg are different transactional table formats with version-dependent feature support. Without this control, assuming every engine can use every feature can result in invalid writes or unreadable metadata. A direct acceptance check is: All required engines pass a controlled read and supported mutation matrix before production.
Incorrect Answers
B:This alternative primarily addresses a different Azure Databricks problem: A large Delta fact table is filtered primarily on high-cardinality account IDs with changing query patterns. The corresponding measure there would be: Enable supported liquid clustering with appropriate keys and run compatible optimization. The stated situation instead calls for: Choose and document a supported format and catalog integration with explicit read/write compatibility tests.
C:This alternative primarily addresses a different Azure Databricks problem: A raw event lake contains append-only records but a developer chooses daily full overwrite of the curated Delta table. The corresponding measure there would be: Choose incremental append or merge based on immutable versus mutable source behavior. The stated situation instead calls for: Choose and document a supported format and catalog integration with explicit read/write compatibility tests.
D:This alternative primarily addresses a different Azure Databricks problem: A pipeline uses CREATE TABLE AS SELECT for a one-time derived dataset but later assumes the result refreshes automatically. The corresponding measure there would be: Select CTAS for an explicitly materialized snapshot or a scheduled pipeline for ongoing refresh. The stated situation instead calls for: Choose and document a supported format and catalog integration with explicit read/write compatibility tests.
Question 32
For this Azure Databricks issue: A bulk repair executes all pipeline tasks again even when only the final reporting notebook failed. Engineers choose: Separate tasks and use the supported repair-run workflow at the failed stage when dependencies remain valid. What acceptance evidence is BEST?
- A staging and production run use the same code with distinct approved configuration and no credential leakage.
- After a final-task failure, a repair completes the target without reprocessing validated upstream data.
- Invalid input blocks publication while the previous approved dataset remains available.
- Timing evidence shows independent branches run concurrently and the merge waits for both results.
Explanation
Correct Answer
B:The strongest acceptance proof is: After a final-task failure, a repair completes the target without reprocessing validated upstream data. It specifically verifies: Separate tasks and use the supported repair-run workflow at the failed stage when dependencies remain valid. The original issue was: A bulk repair executes all pipeline tasks again even when only the final reporting notebook failed. The underlying mechanism is: Task-level repair reduces unnecessary recomputation while preserving dependency correctness. Without it, full reruns can repeat expensive historical ingestion and duplicate side effects.
Incorrect Answers
A:This result would validate a different measure: Parameterize environment-specific paths, catalogs and identities through governed deployment configuration. That is relevant to the distinct problem: A multi-team notebook uses hard-coded workspace names and credentials, making the same pipeline fail in staging. Here the acceptance criterion is: After a final-task failure, a repair completes the target without reprocessing validated upstream data.
C:This result would validate a different measure: Place validations before publishing and use an atomic cutover or controlled target write. That is relevant to the distinct problem: A workflow writes to production tables before a validation task examines null keys and duplicates. Here the acceptance criterion is: After a final-task failure, a repair completes the target without reprocessing validated upstream data.
D:This result would validate a different measure: Design a task DAG with parallel independent ingestion and an aggregation dependency after both complete. That is relevant to the distinct problem: A pipeline uses a single notebook to read four sources sequentially even though two can be processed independently. Here the acceptance criterion is: After a final-task failure, a repair completes the target without reprocessing validated upstream data.
Question 33
A pipeline owner declines the action: Use a snapshot replacement or keyed MERGE strategy instead of appending every full snapshot as new records. The business situation is: A daily file load uses append-only mode for periodic snapshots of the same full inventory. What specific failure can follow?
- Appending full inventories multiplies items and distorts on-hand quantities.
- Daily batches cannot satisfy short update deadlines even when their processing is correct.
- A failed data-quality rule cannot be traced back to the event-bus record that produced it.
- Misparsed columns can populate nulls while the batch appears to complete successfully.
Explanation
Correct Answer
A:The direct residual exposure is: Appending full inventories multiplies items and distorts on-hand quantities. The triggering condition is: A daily file load uses append-only mode for periodic snapshots of the same full inventory. The omitted safeguard is: Use a snapshot replacement or keyed MERGE strategy instead of appending every full snapshot as new records. This matters because: Input extraction semantics determine whether repeated rows are new events or refreshed state. A corrective test would show: Two full snapshots produce one current inventory row per item and expected updates.
Incorrect Answers
B:This failure mode belongs mainly to another situation: A team selects batch processing for customer account changes though downstream applications require minute-scale freshness. Its preventative action would be: Use supported streaming or sufficiently frequent incremental CDC with measured end-to-end latency. The most direct exposure in the current scenario is: Appending full inventories multiplies items and distorts on-hand quantities.
C:This failure mode belongs mainly to another situation: A pipeline ingests Event Hubs messages into a bronze table but never captures the partition, offset or event timestamp. Its preventative action would be: Persist necessary source metadata and checkpoint lineage for troubleshooting and replay. The most direct exposure in the current scenario is: Appending full inventories multiplies items and distorts on-hand quantities.
D:This failure mode belongs mainly to another situation: An overnight load reads tab-separated files but an engineer assumes comma separators and casts malformed columns to null. Its preventative action would be: Specify the correct input delimiter, header and schema and validate parsing on a representative file. The most direct exposure in the current scenario is: Appending full inventories multiplies items and distorts on-hand quantities.
Question 34
A solution applies: Define fact-table grain at the required transaction or line-item level before aggregation. It addresses: A facts table groups product sales at daily level although auditors must inspect individual transaction adjustments. Which operational outcome MOST directly verifies success?
- Sample invoice lines reconcile exactly to the fact table and roll up to the approved daily total.
- Two views return consistent key rows while their documented alias columns remain stable.
- A backdated invoice joins to the address valid on its transaction date while the current record shows the latest address.
- A measured workload benchmark shows scan reduction across the highest-value query templates.
Explanation
Correct Answer
A:The strongest acceptance proof is: Sample invoice lines reconcile exactly to the fact table and roll up to the approved daily total. It specifically verifies: Define fact-table grain at the required transaction or line-item level before aggregation. The original issue was: A facts table groups product sales at daily level although auditors must inspect individual transaction adjustments. The underlying mechanism is: Fact grain determines which analyses can be reconstructed without information loss. Without it, premature aggregation can make returns, corrections and invoice-line disputes impossible to reconcile.
Incorrect Answers
B:This result would validate a different measure: Expose stable SQL views over a governed base table when only presentation aliases differ. That is relevant to the distinct problem: A data product creates multiple physical copies of a curated dimension because teams need different column names. Here the acceptance criterion is: Sample invoice lines reconcile exactly to the fact table and roll up to the approved daily total.
C:This result would validate a different measure: Model a slowly changing dimension Type 2 with effective dates, current flags and correctly managed surrogate keys. That is relevant to the distinct problem: A retail customer table changes delivery address repeatedly but analysts must reconstruct the address known at any prior invoice date. Here the acceptance criterion is: Sample invoice lines reconcile exactly to the fact table and roll up to the approved daily total.
D:This result would validate a different measure: Compare selective filter patterns and data-skipping statistics before committing partition and clustering keys. That is relevant to the distinct problem: A query repeatedly filters by customer, date and product but the physical data design was selected without query-profile evidence. Here the acceptance criterion is: Sample invoice lines reconcile exactly to the fact table and roll up to the approved daily total.
Question 35
During a lakehouse implementation review, engineers observe: A customer-name cleansing script uses an arbitrary lowercase conversion without preserving Unicode and leading/trailing whitespace rules. What is the BEST next action?
- Use an incremental append pattern with suitable deduplication when source events cannot change.
- Define and test consistent trimming, Unicode normalization and case comparison for the business key.
- Use an appropriate intersect or semi-join operation based on set and duplicate requirements.
- Evaluate a suitable broadcast join based on measured dimension size and Spark execution plans.
Explanation
Correct Answer
B:The most appropriate action is: Define and test consistent trimming, Unicode normalization and case comparison for the business key. It addresses the stated condition: A customer-name cleansing script uses an arbitrary lowercase conversion without preserving Unicode and leading/trailing whitespace rules. The technical basis is: Text normalization must be deterministic and tailored to identifier semantics. Without this control, inconsistent name normalization can create duplicate customers or merge distinct identities incorrectly. A direct acceptance check is: Mixed-case, whitespace and non-ASCII test records resolve according to the documented mapping policy.
Incorrect Answers
A:This alternative primarily addresses a different Azure Databricks problem: An append-only telemetry feed includes newly observed events but engineers use MERGE with expensive full-table scans on every insert. The corresponding measure there would be: Use an incremental append pattern with suitable deduplication when source events cannot change. The stated situation instead calls for: Define and test consistent trimming, Unicode normalization and case comparison for the business key.
C:This alternative primarily addresses a different Azure Databricks problem: A team wants only customer IDs present in both systems but uses an inner join that multiplies repeated rows. The corresponding measure there would be: Use an appropriate intersect or semi-join operation based on set and duplicate requirements. The stated situation instead calls for: Define and test consistent trimming, Unicode normalization and case comparison for the business key.
D:This alternative primarily addresses a different Azure Databricks problem: A pipeline joins a huge fact table with a tiny dimension and a normal shuffle join creates excessive network traffic. The corresponding measure there would be: Evaluate a suitable broadcast join based on measured dimension size and Spark execution plans. The stated situation instead calls for: Define and test consistent trimming, Unicode normalization and case comparison for the business key.
Question 36
For this Microsoft DP-750 case: An ad hoc machine-learning preparation task has complex custom Python orchestration and no declarative table graph. Which design or operational response BEST fits?
- Add bounded retries with exponential backoff and observable failure after the retry budget.
- Declare consistent table names and dependencies and validate the generated pipeline graph.
- Separate tasks and use the supported repair-run workflow at the failed stage when dependencies remain valid.
- Select a notebook task within Lakeflow Jobs when procedural control and custom libraries fit the requirements.
Explanation
Correct Answer
D:The most appropriate action is: Select a notebook task within Lakeflow Jobs when procedural control and custom libraries fit the requirements. It addresses the stated condition: An ad hoc machine-learning preparation task has complex custom Python orchestration and no declarative table graph. The technical basis is: Notebooks and declarative pipelines serve different orchestration and transformation patterns. Without this control, forcing arbitrary procedural logic into a declarative table pipeline may complicate testing and maintenance. A direct acceptance check is: A job executes the procedural notebook under controlled dependencies and produces the validated output.
Incorrect Answers
A:This alternative primarily addresses a different Azure Databricks problem: A batch enrichment task downloads a remote file and sometimes receives transient HTTP 503 responses. The corresponding measure there would be: Add bounded retries with exponential backoff and observable failure after the retry budget. The stated situation instead calls for: Select a notebook task within Lakeflow Jobs when procedural control and custom libraries fit the requirements.
B:This alternative primarily addresses a different Azure Databricks problem: A declarative pipeline transforms several datasets but its source names are ambiguous and a data team cannot reconstruct dependencies. The corresponding measure there would be: Declare consistent table names and dependencies and validate the generated pipeline graph. The stated situation instead calls for: Select a notebook task within Lakeflow Jobs when procedural control and custom libraries fit the requirements.
C:This alternative primarily addresses a different Azure Databricks problem: A bulk repair executes all pipeline tasks again even when only the final reporting notebook failed. The corresponding measure there would be: Separate tasks and use the supported repair-run workflow at the failed stage when dependencies remain valid. The stated situation instead calls for: Select a notebook task within Lakeflow Jobs when procedural control and custom libraries fit the requirements.
Question 37
A team omits the recommended measure: Use a managed Delta table for transactional updates and controlled schema evolution. Its environment shows: A reporting team stores frequent updates in raw CSV files and expects atomic MERGE and historical time travel. What is the MOST direct residual risk?
- Directly rewriting CSV in place can expose partial updates and inconsistent reader results.
- Without recoverable raw inputs the team cannot replay a transformation after discovering a rule defect.
- Duplicate dimension keys can inflate sums when each fact joins more than one dimension row.
- Using Type 1 updates may silently rewrite the apparent address on older transaction reports.
Explanation
Correct Answer
A:The direct residual exposure is: Directly rewriting CSV in place can expose partial updates and inconsistent reader results. The triggering condition is: A reporting team stores frequent updates in raw CSV files and expects atomic MERGE and historical time travel. The omitted safeguard is: Use a managed Delta table for transactional updates and controlled schema evolution. This matters because: Delta tables support transactional mutations and versioned metadata beyond raw delimiter files. A corrective test would show: A MERGE transaction updates the expected records atomically and a historical version remains queryable within retention.
Incorrect Answers
B:This failure mode belongs mainly to another situation: A bronze ingestion layer stores only final transformed values and loses raw source fields needed for incident replay. Its preventative action would be: Keep a governed raw or bronze representation before applying irreversible cleansing into silver and gold. The most direct exposure in the current scenario is: Directly rewriting CSV in place can expose partial updates and inconsistent reader results.
C:This failure mode belongs mainly to another situation: A finance dashboard needs monthly revenue by customer and must join customer dimensions without multiplying facts. Its preventative action would be: Design a conformed star schema with explicit fact grain and unique dimension keys. The most direct exposure in the current scenario is: Directly rewriting CSV in place can expose partial updates and inconsistent reader results.
D:This failure mode belongs mainly to another situation: A retail customer table changes delivery address repeatedly but analysts must reconstruct the address known at any prior invoice date. Its preventative action would be: Model a slowly changing dimension Type 2 with effective dates, current flags and correctly managed surrogate keys. The most direct exposure in the current scenario is: Directly rewriting CSV in place can expose partial updates and inconsistent reader results.
Question 38
For this Azure Databricks issue: A Databricks bundle deployment differs from what appears in the repository because someone edits the job in the UI afterward. Engineers choose: Use drift detection or a release reconciliation that compares deployed objects against the versioned bundle. What acceptance evidence is BEST?
- A test UI change is flagged and the approved deployment restores the intended configuration.
- CI deploys under the named service principal and the personal token is rejected and revoked.
- A product owner signs off a sample report and its reconciled acceptance criteria before rollout.
- Deployment plans show the intended trigger for each target and a test job fires at the correct interval.
Explanation
Correct Answer
A:The strongest acceptance proof is: A test UI change is flagged and the approved deployment restores the intended configuration. It specifically verifies: Use drift detection or a release reconciliation that compares deployed objects against the versioned bundle. The original issue was: A Databricks bundle deployment differs from what appears in the repository because someone edits the job in the UI afterward. The underlying mechanism is: Source control only provides operational truth when actual resource settings match the approved definition. Without it, manual post-deployment edits can bypass review and silently change retry or security settings.
Incorrect Answers
B:This result would validate a different measure: Use a secure service identity and protected credential handling for bundle or REST API deployment. That is relevant to the distinct problem: A CI pipeline deploys Databricks assets under an engineer's personal token stored in repository variables. Here the acceptance criterion is: A test UI change is flagged and the approved deployment restores the intended configuration.
C:This result would validate a different measure: Add user acceptance tests for the relevant reporting contract before production promotion. That is relevant to the distinct problem: A developer changes a transformation with a user-visible business effect but omits acceptance testing by the data product owner. Here the acceptance criterion is: A test UI change is flagged and the approved deployment restores the intended configuration.
D:This result would validate a different measure: Validate per-target bundle variable resolution and compare rendered job definitions before deployment. That is relevant to the distinct problem: A data engineer updates a bundled job trigger but another environment's deployment silently uses the wrong schedule. Here the acceptance criterion is: A test UI change is flagged and the approved deployment restores the intended configuration.
Question 39
The design team proposes: Replace broad grants with documented minimum creation and use privileges in a dedicated schema. The situation is: A development team grants ALL PRIVILEGES on a catalog so that one job can create a temporary table. Which technical principle justifies the decision?
- Attribute-based policies can apply common decisions at scale rather than managing every table individually.
- Separation of duties requires independent decisions for object administration, data modification and read access.
- Secret retrieval must be paired with permissions and safe handling to avoid disclosure after retrieval.
- Cloud IAM and Unity Catalog are complementary authorization layers that must both permit the intended operation.
Explanation
Correct Answer
B:The decisive mechanism is: Separation of duties requires independent decisions for object administration, data modification and read access. That explains the selected action: Replace broad grants with documented minimum creation and use privileges in a dedicated schema. In the original scenario: A development team grants ALL PRIVILEGES on a catalog so that one job can create a temporary table. The specific unmanaged risk is: A compromised job principal could alter data products unrelated to its intended table. Validate with: An access test creates the required table only inside the approved job schema and denies other catalogs.
Incorrect Answers
A:This principle concerns a separate data-engineering condition: A data compliance team needs consistent redaction of columns tagged as personal data across hundreds of catalog tables. Its targeted response is: Use centrally governed tags and compatible Unity Catalog ABAC masking policies where supported. The current case is governed more directly by: Separation of duties requires independent decisions for object administration, data modification and read access.
C:This principle concerns a separate data-engineering condition: An application uses an external Azure Key Vault secret scope but a notebook prints the secret into task logs. Its targeted response is: Use supported Databricks secret access with least-privilege scope access and remove secret values from logs. The current case is governed more directly by: Separation of duties requires independent decisions for object administration, data modification and read access.
D:This principle concerns a separate data-engineering condition: A managed identity can read a storage container but a newly created external table is denied in Databricks. Its targeted response is: Align Azure storage role assignments with the registered Unity Catalog storage credential and external-location privileges. The current case is governed more directly by: Separation of duties requires independent decisions for object administration, data modification and read access.
Question 40
The team chooses: Choose incremental append or merge based on immutable versus mutable source behavior. Its observed issue is: A raw event lake contains append-only records but a developer chooses daily full overwrite of the curated Delta table. What is the strongest technical basis?
- Delta tables support transactional mutations and versioned metadata beyond raw delimiter files.
- Temporal tables describe the validity of state over time rather than only its latest value.
- Liquid clustering and traditional ZORDER/partitioning have compatibility constraints and different maintenance paths.
- Ingestion strategy should match whether source records can be corrected or deleted.
Explanation
Correct Answer
D:The decisive mechanism is: Ingestion strategy should match whether source records can be corrected or deleted. That explains the selected action: Choose incremental append or merge based on immutable versus mutable source behavior. In the original scenario: A raw event lake contains append-only records but a developer chooses daily full overwrite of the curated Delta table. The specific unmanaged risk is: Unnecessary overwrites increase compute cost and can cause data loss during partial reruns. Validate with: Two successive batches load only new records while preserved history and counts reconcile.
Incorrect Answers
A:This principle concerns a separate data-engineering condition: A reporting team stores frequent updates in raw CSV files and expects atomic MERGE and historical time travel. Its targeted response is: Use a managed Delta table for transactional updates and controlled schema evolution. The current case is governed more directly by: Ingestion strategy should match whether source records can be corrected or deleted.
B:This principle concerns a separate data-engineering condition: A historical compliance table must show how account status changed at precise effective intervals. Its targeted response is: Use a temporal history model with nonoverlapping effective-from and effective-to boundaries. The current case is governed more directly by: Ingestion strategy should match whether source records can be corrected or deleted.
C:This principle concerns a separate data-engineering condition: A team attempts to apply both ZORDER and liquid clustering to the same Delta table. Its targeted response is: Choose one compatible data layout strategy based on table format, platform support and query evidence. The current case is governed more directly by: Ingestion strategy should match whether source records can be corrected or deleted.