Certified Data Engineer Associate Practice Test 22
Question 210
At a city utilities office, bronze shipping events use mixed timezone strings and inconsistent carrier codes, making downstream shipment durations misleading. The dataset `booking_changes` is in scope, and a finance audit requires reproducible aggregated figures. What verification provides evidence of correct behavior under the original adverse condition?
- Check that the intended configuration exists for `booking_changes` and that a healthy run succeeded, without inducing the earlier failure.
- Run only a normal, unchanged `booking_changes` sample and review timestamp offsets.
- Replay two events at the same instant with different offsets and verify equal canonical time plus mapped carrier identity
- Review a platform-level availability report for `booking_changes` without testing whether heterogeneous timestamps and code names cannot be compared or joined consistently without standardization.
Explanation
Correct Answer
C:Replay two events at the same instant with different offsets and verify equal canonical time plus mapped carrier identity. This reproduces the scenario-dependent condition, verifies timestamp offsets, normalized UTC values, reference lookups, and rejected code counts, and directly tests whether heterogeneous timestamps and code names cannot be compared or joined consistently without standardization is resolved. Observed results, not intent, determine approval.
Incorrect Answers
A:Check that the intended configuration exists for `booking_changes` and that a healthy run succeeded, without inducing the earlier failure. Installed configuration and a healthy run are insufficient to prove the failure mode is gone. The original issue was heterogeneous timestamps and code names cannot be compared or joined consistently without standardization.
B:Run only a normal, unchanged `booking_changes` sample and review timestamp offsets. A normal sample does not challenge heterogeneous timestamps and code names cannot be compared or joined consistently without standardization.
D:Review a platform-level availability report for `booking_changes` without testing whether heterogeneous timestamps and code names cannot be compared or joined consistently without standardization. Aggregate availability does not establish the behavior of affected `booking_changes` records. It cannot replace the direct exercise: replay two events at the same instant with different offsets and verify equal canonical time plus mapped carrier identity.
Question 211
At a ticketing marketplace, bronze orders arrive with currency totals as strings and null order dates, producing incorrect daily financial summaries. The dataset `meter_readings` is in scope, and the release must preserve the original source evidence. Which investigation best tests the stated constraint with direct data?
- Inspect normal-period success metrics for the stated dataset without reproducing the source error
- Review completed run status and aggregate cluster health without record-level correlation
- Correlate source field types, invalid-row sample across affected and unaffected `R034-CASE-043` runs.
- Compare the saved configuration with the signed change ticket for `meter_readings` while omitting source and output records.
Explanation
Correct Answer
C:Correlate source field types, invalid-row sample across affected and unaffected `R034-CASE-043` runs.. By joining source field types, invalid-row sample, cast success count, and silver numeric reconciliation across the affected and unaffected conditions, the investigation directly tests the recorded mechanism: string amounts can aggregate incorrectly and missing business dates are not safe grouping keys. It also sets a measurable baseline for load zero, null and malformed amounts, then reconcile silver totals and quarantined rejects to bronze input.
Incorrect Answers
A:Inspect normal-period success metrics for the stated dataset without reproducing the source error. A healthy-run trend can be useful, but it does not expose string amounts can aggregate incorrectly and missing business dates are not safe grouping keys. The failing-path source field types, invalid-row sample, cast success count, and silver numeric reconciliation has to be correlated before attributing cause.
B:Review completed run status and aggregate cluster health without record-level correlation. A successful aggregate platform state can coexist with string amounts can aggregate incorrectly and missing business dates are not safe grouping keys. This check never observes the affected `meter_readings` records or their causal path.
D:Compare the saved configuration with the signed change ticket for `meter_readings` while omitting source and output records. A desired configuration and approval ticket are not runtime proof. In this case they omit source field types, invalid-row sample, cast success count, and silver numeric reconciliation, the direct means of distinguishing string amounts can aggregate incorrectly and missing business dates are not safe grouping keys.
Question 212
At a ticketing marketplace, bronze orders arrive with currency totals as strings and null order dates, producing incorrect daily financial summaries. The dataset `meter_readings` is in scope, and the release must preserve the original source evidence. Which option targets the actual technical dependency rather than an incidental symptom?
- Validate and cast amounts to a numeric type, quarantine invalid dates, and write typed clean records to a silver Delta table
- Overwrite the original bronze payload with only accepted rows; use that approach for `meter_readings`.
- Replace every missing date with today without provenance; use that approach for `meter_readings`.
- Aggregate the raw string amount without conversion and skip null checks; use that approach for `meter_readings`.
Explanation
Correct Answer
A:Validate and cast amounts to a numeric type, quarantine invalid dates, and write typed clean records to a silver Delta table. This directly addresses string amounts can aggregate incorrectly and missing business dates are not safe grouping keys. The design should be accepted only after load zero, null and malformed amounts, then reconcile silver totals and quarantined rejects to bronze input; the related operational boundary is to preserve immutable raw events but enforce typed, validated silver records for consumption.
Incorrect Answers
B:Overwrite the original bronze payload with only accepted rows; use that approach for `meter_readings`. Destructive cleaning loses the raw evidence needed to replay and audit rejected records. For `meter_readings`, this cannot remove the causal limitation: string amounts can aggregate incorrectly and missing business dates are not safe grouping keys.
C:Replace every missing date with today without provenance; use that approach for `meter_readings`. Inventing transaction dates changes business facts and misstates historical revenue. For `meter_readings`, this cannot remove the causal limitation: string amounts can aggregate incorrectly and missing business dates are not safe grouping keys.
D:Aggregate the raw string amount without conversion and skip null checks; use that approach for `meter_readings`. String comparison or implicit casts can corrupt arithmetic and let invalid dates pass. For `meter_readings`, this cannot remove the causal limitation: string amounts can aggregate incorrectly and missing business dates are not safe grouping keys.
Question 213
At a ticketing marketplace, bronze orders arrive with currency totals as strings and null order dates, producing incorrect daily financial summaries. The dataset `meter_readings` is in scope, and the release must preserve the original source evidence. A teammate proposes: "Overwrite the original bronze payload with only accepted rows." Which technical limitation is the decisive reason not to approve that specific proposal?
- String comparison or implicit casts can corrupt arithmetic and let invalid dates pass. This would apply to an alternative change for `meter_readings`.
- Inventing transaction dates changes business facts and misstates historical revenue. This would apply to another option for `meter_readings`.
- Destructive cleaning loses the raw evidence needed to replay and audit rejected records. The proposal does not repair `meter_readings`.
- The dashboard cached the prior result; no DataFrame operation or underlying data requires correction.
Explanation
Correct Answer
C:Destructive cleaning loses the raw evidence needed to replay and audit rejected records. The proposal does not repair `meter_readings`. This identifies the known limitation of the exact proposed action. Destructive cleaning loses the raw evidence needed to replay and audit rejected records. The underlying problem remains string amounts can aggregate incorrectly and missing business dates are not safe grouping keys; Validate and cast amounts to a numeric type, quarantine invalid dates, and write typed clean records to a silver Delta table instead targets that failure.
Incorrect Answers
A:String comparison or implicit casts can corrupt arithmetic and let invalid dates pass. This would apply to an alternative change for `meter_readings`. This failure applies to the separate proposal "Aggregate the raw string amount without conversion and skip null checks".
B:Inventing transaction dates changes business facts and misstates historical revenue. This would apply to another option for `meter_readings`. This limitation belongs to the different alternative "Replace every missing date with today without provenance". The stem calls for the specific failure of "Overwrite the original bronze payload with only accepted rows".
D:The dashboard cached the prior result; no DataFrame operation or underlying data requires correction. The case provides a concrete failing data or control path; a mere alternate theory does not address string amounts can aggregate incorrectly and missing business dates are not safe grouping keys.
Question 214
At a ticketing marketplace, bronze orders arrive with currency totals as strings and null order dates, producing incorrect daily financial summaries. The dataset `meter_readings` is in scope, and the release must preserve the original source evidence. What operational boundary should be documented for approval?
- Aggregate the raw string amount without conversion and skip null checks
- Overwrite the original bronze payload with only accepted rows
- Preserve immutable raw events but enforce typed, validated silver records for consumption
- Replace every missing date with today without provenance; monitor results only after launch of `meter_readings`.
Explanation
Correct Answer
C:Preserve immutable raw events but enforce typed, validated silver records for consumption. This explicitly bounds the compromise while still addressing string amounts can aggregate incorrectly and missing business dates are not safe grouping keys. Acceptance relies on load zero, null and malformed amounts, then reconcile silver totals and quarantined rejects to bronze input rather than simply declaring the change complete.
Incorrect Answers
A:Aggregate the raw string amount without conversion and skip null checks. String comparison or implicit casts can corrupt arithmetic and let invalid dates pass. Monitoring after deployment does not make this approach a safe bounded tradeoff for `meter_readings`.
B:Overwrite the original bronze payload with only accepted rows. Destructive cleaning loses the raw evidence needed to replay and audit rejected records. Monitoring after deployment does not make this approach a safe bounded tradeoff for `meter_readings`.
D:Replace every missing date with today without provenance; monitor results only after launch of `meter_readings`. Inventing transaction dates changes business facts and misstates historical revenue. Monitoring after deployment does not make this approach a safe bounded tradeoff for `meter_readings`.
Question 215
At a ticketing marketplace, bronze orders arrive with currency totals as strings and null order dates, producing incorrect daily financial summaries. The dataset `meter_readings` is in scope, and the release must preserve the original source evidence. Which validation should be required in the release record?
- Review a platform-level availability report for `meter_readings` without testing whether string amounts can aggregate incorrectly and missing business dates are not safe grouping keys.
- Run only a normal, unchanged `meter_readings` sample and review source field types.
- Load zero, null and malformed amounts, then reconcile silver totals and quarantined rejects to bronze input
- Check that the intended configuration exists for `meter_readings` and that a healthy run succeeded, without inducing the earlier failure.
Explanation
Correct Answer
C:Load zero, null and malformed amounts, then reconcile silver totals and quarantined rejects to bronze input. This reproduces the scenario-dependent condition, verifies source field types, invalid-row sample, cast success count, and silver numeric reconciliation, and directly tests whether string amounts can aggregate incorrectly and missing business dates are not safe grouping keys is resolved. Observed results, not intent, determine approval.
Incorrect Answers
A:Review a platform-level availability report for `meter_readings` without testing whether string amounts can aggregate incorrectly and missing business dates are not safe grouping keys. Aggregate availability does not establish the behavior of affected `meter_readings` records. It cannot replace the direct exercise: load zero, null and malformed amounts, then reconcile silver totals and quarantined rejects to bronze input.
B:Run only a normal, unchanged `meter_readings` sample and review source field types. A normal sample does not challenge string amounts can aggregate incorrectly and missing business dates are not safe grouping keys.
D:Check that the intended configuration exists for `meter_readings` and that a healthy run succeeded, without inducing the earlier failure. Installed configuration and a healthy run are insufficient to prove the failure mode is gone. The original issue was string amounts can aggregate incorrectly and missing business dates are not safe grouping keys.
Question 216
At a regional banking group, a large fact DataFrame is joined to a very small reference dimension but shuffle volume dominates runtime. The dataset `prescription_claims` is in scope, and raw records must remain available for replay. Before choosing a remedy, which investigation actually distinguishes the failure mechanism from a green dashboard?
- Correlate reference table size, Spark physical plan across affected and unaffected `R034-CASE-044` runs.
- Inspect normal-period success metrics for the stated dataset without reproducing the source error
- Review completed run status and aggregate cluster health without record-level correlation
- Compare the current configuration and the signed change approval, without source-row evidence
Explanation
Correct Answer
A:Correlate reference table size, Spark physical plan across affected and unaffected `R034-CASE-044` runs.. By joining reference table size, Spark physical plan, shuffle bytes, and joined row cardinality across the affected and unaffected conditions, the investigation directly tests the recorded mechanism: shuffling both sides is unnecessary when the small reference fits within safe broadcast memory. It also sets a measurable baseline for compare join plans and row counts with and without broadcasting the small dimension.
Incorrect Answers
B:Inspect normal-period success metrics for the stated dataset without reproducing the source error. A healthy-run trend can be useful, but it does not expose shuffling both sides is unnecessary when the small reference fits within safe broadcast memory. The failing-path reference table size, Spark physical plan, shuffle bytes, and joined row cardinality has to be correlated before attributing cause.
C:Review completed run status and aggregate cluster health without record-level correlation. A successful aggregate platform state can coexist with shuffling both sides is unnecessary when the small reference fits within safe broadcast memory. This check never observes the affected `prescription_claims` records or their causal path.
D:Compare the current configuration and the signed change approval, without source-row evidence. A desired configuration and approval ticket are not runtime proof. In this case they omit reference table size, Spark physical plan, shuffle bytes, and joined row cardinality, the direct means of distinguishing shuffling both sides is unnecessary when the small reference fits within safe broadcast memory.
Question 217
At a regional banking group, a large fact DataFrame is joined to a very small reference dimension but shuffle volume dominates runtime. The dataset `prescription_claims` is in scope, and raw records must remain available for replay. Which implementation best resolves the stated defect without breaking the business constraint?
- Broadcast the multi-terabyte fact table instead; use that approach for `prescription_claims`.
- Broadcast the genuinely small dimension and join on the validated business key, checking output cardinality
- Increase shuffle partitions without considering join distribution; use that approach for `prescription_claims`.
- Use a cross join and filter the matches after multiplication; use that approach for `prescription_claims`.
Explanation
Correct Answer
B:Broadcast the genuinely small dimension and join on the validated business key, checking output cardinality. This directly addresses shuffling both sides is unnecessary when the small reference fits within safe broadcast memory. The design should be accepted only after compare join plans and row counts with and without broadcasting the small dimension; the related operational boundary is to use a targeted broadcast when the reference is safely small, not an unconditional broadcast policy.
Incorrect Answers
A:Broadcast the multi-terabyte fact table instead; use that approach for `prescription_claims`. Replicating the large side exceeds executor memory and can cause instability. For `prescription_claims`, this cannot remove the causal limitation: shuffling both sides is unnecessary when the small reference fits within safe broadcast memory.
C:Increase shuffle partitions without considering join distribution; use that approach for `prescription_claims`. More partitions do not eliminate an avoidable shuffle of the small dimension. For `prescription_claims`, this cannot remove the causal limitation: shuffling both sides is unnecessary when the small reference fits within safe broadcast memory.
D:Use a cross join and filter the matches after multiplication; use that approach for `prescription_claims`. A Cartesian product creates excessive intermediate rows and work. For `prescription_claims`, this cannot remove the causal limitation: shuffling both sides is unnecessary when the small reference fits within safe broadcast memory.
Question 218
At a regional banking group, a large fact DataFrame is joined to a very small reference dimension but shuffle volume dominates runtime. The dataset `prescription_claims` is in scope, and raw records must remain available for replay. A teammate proposes: "Increase shuffle partitions without considering join distribution." Which technical limitation is the decisive reason not to approve that specific proposal?
- Replicating the large side exceeds executor memory and can cause instability. This would apply to an alternative change for `prescription_claims`.
- A Cartesian product creates excessive intermediate rows and work. This would apply to another option for `prescription_claims`.
- More partitions do not eliminate an avoidable shuffle of the small dimension. The proposal does not repair `prescription_claims`.
- The dashboard cached the prior result; no DataFrame operation or underlying data requires correction.
Explanation
Correct Answer
C:More partitions do not eliminate an avoidable shuffle of the small dimension. The proposal does not repair `prescription_claims`. This identifies the known limitation of the exact proposed action. More partitions do not eliminate an avoidable shuffle of the small dimension. The underlying problem remains shuffling both sides is unnecessary when the small reference fits within safe broadcast memory; Broadcast the genuinely small dimension and join on the validated business key, checking output cardinality instead targets that failure.
Incorrect Answers
A:Replicating the large side exceeds executor memory and can cause instability. This would apply to an alternative change for `prescription_claims`. This failure applies to the separate proposal "Broadcast the multi-terabyte fact table instead".
B:A Cartesian product creates excessive intermediate rows and work. This would apply to another option for `prescription_claims`. This limitation belongs to the different alternative "Use a cross join and filter the matches after multiplication". The stem calls for the specific failure of "Increase shuffle partitions without considering join distribution".
D:The dashboard cached the prior result; no DataFrame operation or underlying data requires correction. The case provides a concrete failing data or control path; a mere alternate theory does not address shuffling both sides is unnecessary when the small reference fits within safe broadcast memory.
Question 219
At a regional banking group, a large fact DataFrame is joined to a very small reference dimension but shuffle volume dominates runtime. The dataset `prescription_claims` is in scope, and raw records must remain available for replay. Which bounded tradeoff would preserve the core outcome and control operational risk?
- Broadcast the multi-terabyte fact table instead
- Increase shuffle partitions without considering join distribution
- Use a cross join and filter the matches after multiplication
- Use a targeted broadcast when the reference is safely small, not an unconditional broadcast policy
Explanation
Correct Answer
D:Use a targeted broadcast when the reference is safely small, not an unconditional broadcast policy. This explicitly bounds the compromise while still addressing shuffling both sides is unnecessary when the small reference fits within safe broadcast memory. Acceptance relies on compare join plans and row counts with and without broadcasting the small dimension rather than simply declaring the change complete.
Incorrect Answers
A:Broadcast the multi-terabyte fact table instead. Replicating the large side exceeds executor memory and can cause instability. Monitoring after deployment does not make this approach a safe bounded tradeoff for `prescription_claims`.
B:Increase shuffle partitions without considering join distribution. More partitions do not eliminate an avoidable shuffle of the small dimension. Monitoring after deployment does not make this approach a safe bounded tradeoff for `prescription_claims`.
C:Use a cross join and filter the matches after multiplication. A Cartesian product creates excessive intermediate rows and work. Monitoring after deployment does not make this approach a safe bounded tradeoff for `prescription_claims`.
Question 220
At a regional banking group, a large fact DataFrame is joined to a very small reference dimension but shuffle volume dominates runtime. The dataset `prescription_claims` is in scope, and raw records must remain available for replay. Before release, which acceptance test would actually demonstrate that the root problem is controlled?
- Check that the intended configuration exists for `prescription_claims` and that a healthy run succeeded, without inducing the earlier failure.
- Run only a normal, unchanged `prescription_claims` sample and review reference table size.
- Compare join plans and row counts with and without broadcasting the small dimension
- Review a platform-level availability report for `prescription_claims` without testing whether shuffling both sides is unnecessary when the small reference fits within safe broadcast memory.
Explanation
Correct Answer
C:Compare join plans and row counts with and without broadcasting the small dimension. This reproduces the scenario-dependent condition, verifies reference table size, Spark physical plan, shuffle bytes, and joined row cardinality, and directly tests whether shuffling both sides is unnecessary when the small reference fits within safe broadcast memory is resolved. Observed results, not intent, determine approval.
Incorrect Answers
A:Check that the intended configuration exists for `prescription_claims` and that a healthy run succeeded, without inducing the earlier failure. Installed configuration and a healthy run are insufficient to prove the failure mode is gone. The original issue was shuffling both sides is unnecessary when the small reference fits within safe broadcast memory.
B:Run only a normal, unchanged `prescription_claims` sample and review reference table size. A normal sample does not challenge shuffling both sides is unnecessary when the small reference fits within safe broadcast memory.
D:Review a platform-level availability report for `prescription_claims` without testing whether shuffling both sides is unnecessary when the small reference fits within safe broadcast memory. Aggregate availability does not establish the behavior of affected `prescription_claims` records. It cannot replace the direct exercise: compare join plans and row counts with and without broadcasting the small dimension.