100% Latest & Updated Databricks Certified Data Engineer Associate Practice Test Questions, Exam Dumps & Verified Answers!
30 Days Free Updates, Instant Download!
Question 200
A small static country table enriches a huge events DataFrame and a large shuffle must be avoided when eligible. The team observes: Spark shuffles both huge facts and small lookup records despite an eligible broadcast dimension. Which configuration or design gap best explains the symptom?
- Planner thresholds or statistics prevented a beneficial broadcast choice.
- A batch overwrite replaced the persistent incremental streaming target.
- The join planner was not guided toward the small eligible broadcast side.
- A standard view was used where materialization would meet the latency and freshness contract.
Explanation
Correct Answer
C:The join planner was not guided toward the small eligible broadcast side. Apply the supported broadcast hint to the small dimension on a compatible equijoin.
Incorrect Answers
A:If a large fact-to-small-dimension join repeatedly performs an avoidable shuffle, “Planner thresholds or statistics prevented a beneficial broadcast choice” may be the appropriate response. The join planner was not guided toward the small eligible broadcast side.
B:A downstream pipeline needs incremental consumption of a continuous events feed into a governed table. In those circumstances, “A batch overwrite replaced the persistent incremental streaming target” fits. The join planner was not guided toward the small eligible broadcast side.
D:If a busy BI dashboard repeatedly runs an expensive grouped query and tolerates refresh-based staleness, “A standard view was used where materialization would meet the latency and freshness contract” may be the appropriate response. The join planner was not guided toward the small eligible broadcast side.
Question 201
A customer match requires both organization identifier and customer identifier, since a customer number is reused by different businesses. Which architecture or engineering decision best fits this scenario?
- Evaluate the supported input and execution partitioning strategy for available parallelism.
- Join the data on the composite pair of organization_id and customer_id.
- Apply a row filter with a typed comparison before serving the data.
- Validate currencies and join the approved conversion reference before writing standardized silver amounts.
Explanation
Correct Answer
B:Join the data on the composite pair of organization_id and customer_id. A repeated customer number in two organizations never cross-matches records.
Incorrect Answers
A:A Spark job has many tiny input partitions and inefficient parallel task utilization. In those circumstances, “Evaluate the supported input and execution partitioning strategy for available parallelism” fits. Join the data on the composite pair of organization_id and customer_id.
C:If a bronze telemetry table must expose only current readings above a threshold for a dashboard, “Apply a row filter with a typed comparison before serving the data” may be the appropriate response. Join the data on the composite pair of organization_id and customer_id.
D:A marketing bronze feed arrives with mixed currency columns, but conversion rates are supplied in a governed reference table. In those circumstances, “Validate currencies and join the approved conversion reference before writing standardized silver amounts” fits. Join the data on the composite pair of organization_id and customer_id.
Question 202
A customer match requires both organization identifier and customer identifier, since a customer number is reused by different businesses. Which implementation step is most appropriate before production use?
- Apply explode to the array and retain the parent order identifier in each generated row.
- Supply both key equality predicates to the Spark join condition.
- Cast the measurement to numeric and use filter on the validated threshold predicate.
- Join source customers against revoked ids using left_anti on the right key.
Explanation
Correct Answer
B:Supply both key equality predicates to the Spark join condition. A repeated customer number in two organizations never cross-matches records.
Incorrect Answers
A:When a line-item DataFrame contains an array of product codes per order and downstream rows must be per product, “Apply explode to the array and retain the parent order identifier in each generated row” could be relevant. Supply both key equality predicates to the Spark join condition.
C:A bronze telemetry table must expose only current readings above a threshold for a dashboard. In those circumstances, “Cast the measurement to numeric and use filter on the validated threshold predicate” fits. Supply both key equality predicates to the Spark join condition.
D:When a GDPR deletion feed requires excluding rows whose identifiers appear on a revoked-consent list, “Join source customers against revoked ids using left_anti on the right key” could be relevant. Supply both key equality predicates to the Spark join condition.
Question 203
A customer match requires both organization identifier and customer identifier, since a customer number is reused by different businesses. Which acceptance test provides the strongest evidence that the intended outcome has been achieved?
- A crafted negative amount is rejected or quarantined and never reaches gold.
- A repeated customer number in two organizations never cross-matches records.
- A late-arriving older event does not replace a newer valid customer state.
- The accepted plus rejected counts reconcile to the original source batch.
Explanation
Correct Answer
B:A repeated customer number in two organizations never cross-matches records.
Incorrect Answers
A:If a payments silver table must never have negative approved transaction amounts, “A crafted negative amount is rejected or quarantined and never reaches gold” may be the appropriate response. A repeated customer number in two organizations never cross-matches records.
C:When a customer dataset must retain the latest update per business key even when events arrive out of order, “A late-arriving older event does not replace a newer valid customer state” could be relevant. A repeated customer number in two organizations never cross-matches records.
D:If a silver quality gate must reject records missing a transaction identifier, but analysts still need an error count, “The accepted plus rejected counts reconcile to the original source batch” may be the appropriate response. A repeated customer number in two organizations never cross-matches records.
Question 204
A customer match requires both organization identifier and customer identifier, since a customer number is reused by different businesses. The team observes: A result joins unrelated companies that happen to reuse a customer number. Which root cause should be addressed first given this evidence?
- The transformation lacked deterministic timestamp and tie-breaking logic.
- The join predicate used customer_id alone, omitting organization_id.
- A materialized object was not refreshed and should have been a view for this need.
- The partition or memory design exceeds available per-task memory.
Explanation
Correct Answer
B:The join predicate used customer_id alone, omitting organization_id. Supply both key equality predicates to the Spark join condition.
Incorrect Answers
A:A health feed has duplicate late-arrival patient observations but raw arrivals must remain reviewable. In those circumstances, “The transformation lacked deterministic timestamp and tie-breaking logic” fits. The join predicate used customer_id alone, omitting organization_id.
C:If a finance dashboard reads a curated metric that should always reflect the latest base table without storing a separately maintained aggregate, “A materialized object was not refreshed and should have been a view for this need” may be the appropriate response. The join predicate used customer_id alone, omitting organization_id.
D:A large join spills to disk repeatedly although executor memory pressure is high. In those circumstances, “The partition or memory design exceeds available per-task memory” fits. The join predicate used customer_id alone, omitting organization_id.
Question 205
Two quarterly datasets have identical columns but their records must be vertically appended and duplicates preserved. Which technical design is most appropriate for the stated constraint?
- Use conditional aggregation on delivery status rather than counting every record as complete.
- Tune shuffle partition count using observed data volume and task metrics rather than a fixed guess.
- Use a documented data quality constraint with a quarantine or reject accounting strategy.
- Use UNION ALL or DataFrame unionByName rather than a relational join.
Explanation
Correct Answer
D:Use UNION ALL or DataFrame unionByName rather than a relational join. The combined row count equals the sum of both inputs when no filter is applied.
Incorrect Answers
A:When a service operations report must count completed deliveries separately from all attempted deliveries, “Use conditional aggregation on delivery status rather than counting every record as complete” could be relevant. Use UNION ALL or DataFrame unionByName rather than a relational join.
B:If a heavily shuffled ETL join creates thousands of tiny shuffle partitions while tasks spend time scheduling, “Tune shuffle partition count using observed data volume and task metrics rather than a fixed guess” may be the appropriate response. Use UNION ALL or DataFrame unionByName rather than a relational join.
C:A silver quality gate must reject records missing a transaction identifier, but analysts still need an error count. In those circumstances, “Use a documented data quality constraint with a quarantine or reject accounting strategy” fits. Use UNION ALL or DataFrame unionByName rather than a relational join.
Question 206
Two quarterly datasets have identical columns but their records must be vertically appended and duplicates preserved. Which action should the engineer perform to establish the intended control?
- Apply a documented timestamp parser and timezone normalization while writing typed silver columns.
- Select compatible columns and apply unionByName with an explicit schema contract.
- Define a supported materialized view in Unity Catalog and configure refresh according to requirements.
- Apply a nonnull id test and route rejects to an auditable quarantine table.
Explanation
Correct Answer
B:Select compatible columns and apply unionByName with an explicit schema contract. The combined row count equals the sum of both inputs when no filter is applied.
Incorrect Answers
A:If a payroll bronze table has event timestamps saved as arbitrary text from two time zones, “Apply a documented timestamp parser and timezone normalization while writing typed silver columns” may be the appropriate response. Select compatible columns and apply unionByName with an explicit schema contract.
C:When a busy BI dashboard repeatedly runs an expensive grouped query and tolerates refresh-based staleness, “Define a supported materialized view in Unity Catalog and configure refresh according to requirements” could be relevant. Select compatible columns and apply unionByName with an explicit schema contract.
D:If a silver quality gate must reject records missing a transaction identifier, but analysts still need an error count, “Apply a nonnull id test and route rejects to an auditable quarantine table” may be the appropriate response. Select compatible columns and apply unionByName with an explicit schema contract.
Question 207
Two quarterly datasets have identical columns but their records must be vertically appended and duplicates preserved. Which observation best proves the workload meets its requirement?
- Two payload retries with the same event identifier contribute one event to the count.
- The combined row count equals the sum of both inputs when no filter is applied.
- Equivalent instants from different offsets compare equal and raw source text remains available.
- Known valid orders have typed totals and normalized country values, while rejected keyless orders remain traceable to bronze.
Explanation
Correct Answer
B:The combined row count equals the sum of both inputs when no filter is applied.
Incorrect Answers
A:A click event feed contains retries with identical event_id and should count each business event once. In those circumstances, “Two payload retries with the same event identifier contribute one event to the count” fits. The combined row count equals the sum of both inputs when no filter is applied.
C:If a payroll bronze table has event timestamps saved as arbitrary text from two time zones, “Equivalent instants from different offsets compare equal and raw source text remains available” may be the appropriate response. The combined row count equals the sum of both inputs when no filter is applied.
D:A retail bronze orders table contains string totals, inconsistent country codes and missing order identifiers before a silver SLA can begin. In those circumstances, “Known valid orders have typed totals and normalized country values, while rejected keyless orders remain traceable to bronze” fits. The combined row count equals the sum of both inputs when no filter is applied.
Question 208
Two quarterly datasets have identical columns but their records must be vertically appended and duplicates preserved. The team observes: One of two duplicate transactions disappears after combining periods. Which underlying cause most directly explains this failure?
- The job insisted on a full exact DISTINCT despite permitted approximate accuracy.
- The pipeline left the legacy column name or dropped it without a replacement.
- The pipeline used duplicate-eliminating UNION rather than UNION ALL semantics.
- The joining and aggregation grain created many-to-many duplication without reconciliation checks.
Explanation
Correct Answer
C:The pipeline used duplicate-eliminating UNION rather than UNION ALL semantics. Select compatible columns and apply unionByName with an explicit schema contract.
Incorrect Answers
A:When a reporting table must count unique devices across billions of events with an acceptable approximation, “The job insisted on a full exact DISTINCT despite permitted approximate accuracy” could be relevant. The pipeline used duplicate-eliminating UNION rather than UNION ALL semantics.
B:If a customer schema is changing from old_customer_id to customer_id while dashboards expect the new name, “The pipeline left the legacy column name or dropped it without a replacement” may be the appropriate response. The pipeline used duplicate-eliminating UNION rather than UNION ALL semantics.
D:When an insurance gold fact table combines multiple silver sources with different dates and must remain reconcilable, “The joining and aggregation grain created many-to-many duplication without reconciliation checks” could be relevant. The pipeline used duplicate-eliminating UNION rather than UNION ALL semantics.
Question 209
A risk team needs all combinations of five scenario assumptions and six stress categories. Which design choice best addresses the stated business and technical requirement?
- Apply a deliberate cross join to form a bounded Cartesian product.
- Apply a row filter with a typed comparison before serving the data.
- Evaluate the supported input and execution partitioning strategy for available parallelism.
- Project only the explicitly approved fields rather than carrying the full source schema.
Explanation
Correct Answer
A:Apply a deliberate cross join to form a bounded Cartesian product. The result contains exactly thirty combinations with no missing assumption-category pair.
Incorrect Answers
B:A bronze telemetry table must expose only current readings above a threshold for a dashboard. In those circumstances, “Apply a row filter with a typed comparison before serving the data” fits. Apply a deliberate cross join to form a bounded Cartesian product.
C:When a Spark job has many tiny input partitions and inefficient parallel task utilization, “Evaluate the supported input and execution partitioning strategy for available parallelism” could be relevant. Apply a deliberate cross join to form a bounded Cartesian product.
D:If an account table has three irrelevant vendor-internal columns that must not reach the published silver table, “Project only the explicitly approved fields rather than carrying the full source schema” may be the appropriate response. Apply a deliberate cross join to form a bounded Cartesian product.
Question 210
A risk team needs all combinations of five scenario assumptions and six stress categories. Which setup step most directly realizes the proposed design?
- Validate the bounded dimensions and crossJoin the two small frames.
- Partition by customer_id, order by timestamp and event sequence descending, and keep row_number=1.
- Implement an amount >= 0 validation and define an explicit fail or quarantine action.
- Group the relevant rows and apply approx_count_distinct to device_id.
Explanation
Correct Answer
A:Validate the bounded dimensions and crossJoin the two small frames. The result contains exactly thirty combinations with no missing assumption-category pair.
Incorrect Answers
B:When a customer dataset must retain the latest update per business key even when events arrive out of order, “Partition by customer_id, order by timestamp and event sequence descending, and keep row_number=1” could be relevant. Validate the bounded dimensions and crossJoin the two small frames.
C:If a payments silver table must never have negative approved transaction amounts, “Implement an amount >= 0 validation and define an explicit fail or quarantine action” may be the appropriate response. Validate the bounded dimensions and crossJoin the two small frames.
D:A reporting table must count unique devices across billions of events with an acceptable approximation. In those circumstances, “Group the relevant rows and apply approx_count_distinct to device_id” fits. Validate the bounded dimensions and crossJoin the two small frames.
Certified Data Engineer Associate Practice Tests