Certified Data Engineer Associate Practice Test 16
Question 150
At a hospital analytics group, a finance notebook connects to an external JDBC source with credentials embedded directly in shared code. The dataset `payment_decisions` is in scope, and the target must retain every valid late-arriving event. What verification provides evidence of correct behavior under the original adverse condition?
- Review a platform-level availability report for `payment_decisions` without testing whether hard-coded JDBC credentials leak in notebook revisions and can outlive approved rotations.
- Run only a normal, unchanged `payment_decisions` sample and review secret access audit.
- Rotate the database credential and verify the job still retrieves it securely without exposing it in source code
- Check that the intended configuration exists for `payment_decisions` and that a healthy run succeeded, without inducing the earlier failure.
Explanation
Correct Answer
C:Rotate the database credential and verify the job still retrieves it securely without exposing it in source code. This reproduces the scenario-dependent condition, verifies secret access audit, JDBC effective role, notebook revisions, and scheduled job result, and directly tests whether hard-coded JDBC credentials leak in notebook revisions and can outlive approved rotations is resolved. Observed results, not intent, determine approval.
Incorrect Answers
A:Review a platform-level availability report for `payment_decisions` without testing whether hard-coded JDBC credentials leak in notebook revisions and can outlive approved rotations. Aggregate availability does not establish the behavior of affected `payment_decisions` records. It cannot replace the direct exercise: rotate the database credential and verify the job still retrieves it securely without exposing it in source code.
B:Run only a normal, unchanged `payment_decisions` sample and review secret access audit. A normal sample does not challenge hard-coded JDBC credentials leak in notebook revisions and can outlive approved rotations.
D:Check that the intended configuration exists for `payment_decisions` 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 hard-coded JDBC credentials leak in notebook revisions and can outlive approved rotations.
Question 151
At a municipal transit operator, a warehouse receives millions of newly arriving object-storage files each day and a SQL-only full scan has become slow. The dataset `account_statements` is in scope, and an interrupted delivery must be recoverable without data loss. Which investigation best tests the stated constraint with direct data?
- Correlate new-file arrival rate, discovery latency across affected and unaffected `R034-CASE-031` 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 saved configuration with the signed change ticket for `account_statements` while omitting source and output records.
Explanation
Correct Answer
A:Correlate new-file arrival rate, discovery latency across affected and unaffected `R034-CASE-031` runs.. By joining new-file arrival rate, discovery latency, listing overhead, and checkpoint progress across the affected and unaffected conditions, the investigation directly tests the recorded mechanism: listing all historical objects for each run scales poorly under continual file arrival. It also sets a measurable baseline for double the arriving file count and verify sustained ingestion throughput without rereading historical files.
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 listing all historical objects for each run scales poorly under continual file arrival. The failing-path new-file arrival rate, discovery latency, listing overhead, and checkpoint progress 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 listing all historical objects for each run scales poorly under continual file arrival. This check never observes the affected `account_statements` records or their causal path.
D:Compare the saved configuration with the signed change ticket for `account_statements` while omitting source and output records. A desired configuration and approval ticket are not runtime proof. In this case they omit new-file arrival rate, discovery latency, listing overhead, and checkpoint progress, the direct means of distinguishing listing all historical objects for each run scales poorly under continual file arrival.
Question 152
At a municipal transit operator, a warehouse receives millions of newly arriving object-storage files each day and a SQL-only full scan has become slow. The dataset `account_statements` is in scope, and an interrupted delivery must be recoverable without data loss. Which option targets the actual technical dependency rather than an incidental symptom?
- Route the cloud files through an unrelated JDBC database first; use that approach for `account_statements`.
- Favor a scalable Auto Loader or streaming-table ingestion path with file discovery and checkpointing rather than repeated whole-directory scans
- Collect files manually in a notebook driver directory before loading; use that approach for `account_statements`.
- Increase the frequency of complete directory scans without tracking seen files; use that approach for `account_statements`.
Explanation
Correct Answer
B:Favor a scalable Auto Loader or streaming-table ingestion path with file discovery and checkpointing rather than repeated whole-directory scans. This directly addresses listing all historical objects for each run scales poorly under continual file arrival. The design should be accepted only after double the arriving file count and verify sustained ingestion throughput without rereading historical files; the related operational boundary is to accept managed discovery and checkpoint state to meet large-scale freshness requirements.
Incorrect Answers
A:Route the cloud files through an unrelated JDBC database first; use that approach for `account_statements`. An extra relational hop adds complexity without addressing file discovery scale. For `account_statements`, this cannot remove the causal limitation: listing all historical objects for each run scales poorly under continual file arrival.
C:Collect files manually in a notebook driver directory before loading; use that approach for `account_statements`. Driver staging is not scalable or durable for millions of incoming objects. For `account_statements`, this cannot remove the causal limitation: listing all historical objects for each run scales poorly under continual file arrival.
D:Increase the frequency of complete directory scans without tracking seen files; use that approach for `account_statements`. More repeated scans increase object listing cost and duplicate ingestion risk. For `account_statements`, this cannot remove the causal limitation: listing all historical objects for each run scales poorly under continual file arrival.
Question 153
At a municipal transit operator, a warehouse receives millions of newly arriving object-storage files each day and a SQL-only full scan has become slow. The dataset `account_statements` is in scope, and an interrupted delivery must be recoverable without data loss. A teammate proposes: "Route the cloud files through an unrelated JDBC database first." Which technical limitation is the decisive reason not to approve that specific proposal?
- An extra relational hop adds complexity without addressing file discovery scale. The proposal does not repair `account_statements`.
- Insufficient executor memory is the primary fault; source-file handling and ingestion checkpoints are correct.
- More repeated scans increase object listing cost and duplicate ingestion risk. This would apply to another option for `account_statements`.
- Driver staging is not scalable or durable for millions of incoming objects. This would apply to an alternative change for `account_statements`.
Explanation
Correct Answer
A:An extra relational hop adds complexity without addressing file discovery scale. The proposal does not repair `account_statements`. This identifies the known limitation of the exact proposed action. An extra relational hop adds complexity without addressing file discovery scale. The underlying problem remains listing all historical objects for each run scales poorly under continual file arrival; Favor a scalable Auto Loader or streaming-table ingestion path with file discovery and checkpointing rather than repeated whole-directory scans instead targets that failure.
Incorrect Answers
B:Insufficient executor memory is the primary fault; source-file handling and ingestion checkpoints are correct. The case provides a concrete failing data or control path; a mere alternate theory does not address listing all historical objects for each run scales poorly under continual file arrival.
C:More repeated scans increase object listing cost and duplicate ingestion risk. This would apply to another option for `account_statements`. This limitation belongs to the different alternative "Increase the frequency of complete directory scans without tracking seen files". The stem calls for the specific failure of "Route the cloud files through an unrelated JDBC database first".
D:Driver staging is not scalable or durable for millions of incoming objects. This would apply to an alternative change for `account_statements`. This failure applies to the separate proposal "Collect files manually in a notebook driver directory before loading".
Question 154
At a municipal transit operator, a warehouse receives millions of newly arriving object-storage files each day and a SQL-only full scan has become slow. The dataset `account_statements` is in scope, and an interrupted delivery must be recoverable without data loss. What operational boundary should be documented for approval?
- Increase the frequency of complete directory scans without tracking seen files
- Accept managed discovery and checkpoint state to meet large-scale freshness requirements
- Route the cloud files through an unrelated JDBC database first
- Collect files manually in a notebook driver directory before loading; monitor results only after launch of `account_statements`.
Explanation
Correct Answer
B:Accept managed discovery and checkpoint state to meet large-scale freshness requirements. This explicitly bounds the compromise while still addressing listing all historical objects for each run scales poorly under continual file arrival. Acceptance relies on double the arriving file count and verify sustained ingestion throughput without rereading historical files rather than simply declaring the change complete.
Incorrect Answers
A:Increase the frequency of complete directory scans without tracking seen files. More repeated scans increase object listing cost and duplicate ingestion risk. Monitoring after deployment does not make this approach a safe bounded tradeoff for `account_statements`.
C:Route the cloud files through an unrelated JDBC database first. An extra relational hop adds complexity without addressing file discovery scale. Monitoring after deployment does not make this approach a safe bounded tradeoff for `account_statements`.
D:Collect files manually in a notebook driver directory before loading; monitor results only after launch of `account_statements`. Driver staging is not scalable or durable for millions of incoming objects. Monitoring after deployment does not make this approach a safe bounded tradeoff for `account_statements`.
Question 155
At a municipal transit operator, a warehouse receives millions of newly arriving object-storage files each day and a SQL-only full scan has become slow. The dataset `account_statements` is in scope, and an interrupted delivery must be recoverable without data loss. Which validation should be required in the release record?
- Double the arriving file count and verify sustained ingestion throughput without rereading historical files
- Review a platform-level availability report for `account_statements` without testing whether listing all historical objects for each run scales poorly under continual file arrival.
- Run only a normal, unchanged `account_statements` sample and review new-file arrival rate.
- Check that the intended configuration exists for `account_statements` and that a healthy run succeeded, without inducing the earlier failure.
Explanation
Correct Answer
A:Double the arriving file count and verify sustained ingestion throughput without rereading historical files. This reproduces the scenario-dependent condition, verifies new-file arrival rate, discovery latency, listing overhead, and checkpoint progress, and directly tests whether listing all historical objects for each run scales poorly under continual file arrival is resolved. Observed results, not intent, determine approval.
Incorrect Answers
B:Review a platform-level availability report for `account_statements` without testing whether listing all historical objects for each run scales poorly under continual file arrival. Aggregate availability does not establish the behavior of affected `account_statements` records. It cannot replace the direct exercise: double the arriving file count and verify sustained ingestion throughput without rereading historical files.
C:Run only a normal, unchanged `account_statements` sample and review new-file arrival rate. A normal sample does not challenge listing all historical objects for each run scales poorly under continual file arrival.
D:Check that the intended configuration exists for `account_statements` 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 listing all historical objects for each run scales poorly under continual file arrival.
Question 156
At a subscription media service, a supported CRM application offers a managed incremental connector, but the team plans to export full JSON snapshots hourly. The dataset `fraud_signals` is in scope, and the target must retain every valid late-arriving event. Before choosing a remedy, which investigation actually distinguishes the failure mechanism from a green dashboard?
- 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
- Inspect normal-period success metrics for the stated dataset without reproducing the source error
- Correlate CRM connector support matrix, update/delete capture across affected and unaffected `R034-CASE-032` runs.
Explanation
Correct Answer
D:Correlate CRM connector support matrix, update/delete capture across affected and unaffected `R034-CASE-032` runs.. By joining CRM connector support matrix, update/delete capture, replication delay, and target governance across the affected and unaffected conditions, the investigation directly tests the recorded mechanism: custom snapshots add extraction maintenance and cannot reliably detect late deletions. It also sets a measurable baseline for change, delete and re-create a source entity and verify incremental target convergence.
Incorrect Answers
A:Review completed run status and aggregate cluster health without record-level correlation. A successful aggregate platform state can coexist with custom snapshots add extraction maintenance and cannot reliably detect late deletions. This check never observes the affected `fraud_signals` records or their causal path.
B: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 CRM connector support matrix, update/delete capture, replication delay, and target governance, the direct means of distinguishing custom snapshots add extraction maintenance and cannot reliably detect late deletions.
C: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 custom snapshots add extraction maintenance and cannot reliably detect late deletions. The failing-path CRM connector support matrix, update/delete capture, replication delay, and target governance has to be correlated before attributing cause.
Question 157
At a subscription media service, a supported CRM application offers a managed incremental connector, but the team plans to export full JSON snapshots hourly. The dataset `fraud_signals` is in scope, and the target must retain every valid late-arriving event. Which implementation best resolves the stated defect without breaking the business constraint?
- Use Auto Loader to monitor a folder that the CRM never exports into; use that approach for `fraud_signals`.
- Use SQL COPY INTO on the CRM REST URL directly; use that approach for `fraud_signals`.
- Prefer a supported Lakeflow Connect managed connector for source-specific incremental updates and Unity Catalog delivery
- Maintain hourly full snapshots without primary keys; use that approach for `fraud_signals`.
Explanation
Correct Answer
C:Prefer a supported Lakeflow Connect managed connector for source-specific incremental updates and Unity Catalog delivery. This directly addresses custom snapshots add extraction maintenance and cannot reliably detect late deletions. The design should be accepted only after change, delete and re-create a source entity and verify incremental target convergence; the related operational boundary is to choose a managed source integration over custom polling while respecting connector limitations.
Incorrect Answers
A:Use Auto Loader to monitor a folder that the CRM never exports into; use that approach for `fraud_signals`. A file-ingestion service cannot read data that is not delivered as files. For `fraud_signals`, this cannot remove the causal limitation: custom snapshots add extraction maintenance and cannot reliably detect late deletions.
B:Use SQL COPY INTO on the CRM REST URL directly; use that approach for `fraud_signals`. COPY INTO accepts supported file locations, not arbitrary CRM API requests. For `fraud_signals`, this cannot remove the causal limitation: custom snapshots add extraction maintenance and cannot reliably detect late deletions.
D:Maintain hourly full snapshots without primary keys; use that approach for `fraud_signals`. Repeated full snapshots do not guarantee an accurate current-state target or deleted records. For `fraud_signals`, this cannot remove the causal limitation: custom snapshots add extraction maintenance and cannot reliably detect late deletions.
Question 158
At a subscription media service, a supported CRM application offers a managed incremental connector, but the team plans to export full JSON snapshots hourly. The dataset `fraud_signals` is in scope, and the target must retain every valid late-arriving event. A teammate proposes: "Maintain hourly full snapshots without primary keys." Which technical limitation is the decisive reason not to approve that specific proposal?
- Insufficient executor memory is the primary fault; source-file handling and ingestion checkpoints are correct.
- COPY INTO accepts supported file locations, not arbitrary CRM API requests. This would apply to another option for `fraud_signals`.
- Repeated full snapshots do not guarantee an accurate current-state target or deleted records. The proposal does not repair `fraud_signals`.
- A file-ingestion service cannot read data that is not delivered as files. This would apply to an alternative change for `fraud_signals`.
Explanation
Correct Answer
C:Repeated full snapshots do not guarantee an accurate current-state target or deleted records. The proposal does not repair `fraud_signals`. This identifies the known limitation of the exact proposed action. Repeated full snapshots do not guarantee an accurate current-state target or deleted records. The underlying problem remains custom snapshots add extraction maintenance and cannot reliably detect late deletions; Prefer a supported Lakeflow Connect managed connector for source-specific incremental updates and Unity Catalog delivery instead targets that failure.
Incorrect Answers
A:Insufficient executor memory is the primary fault; source-file handling and ingestion checkpoints are correct. The case provides a concrete failing data or control path; a mere alternate theory does not address custom snapshots add extraction maintenance and cannot reliably detect late deletions.
B:COPY INTO accepts supported file locations, not arbitrary CRM API requests. This would apply to another option for `fraud_signals`. This limitation belongs to the different alternative "Use SQL COPY INTO on the CRM REST URL directly". The stem calls for the specific failure of "Maintain hourly full snapshots without primary keys".
D:A file-ingestion service cannot read data that is not delivered as files. This would apply to an alternative change for `fraud_signals`. This failure applies to the separate proposal "Use Auto Loader to monitor a folder that the CRM never exports into".
Question 159
At a subscription media service, a supported CRM application offers a managed incremental connector, but the team plans to export full JSON snapshots hourly. The dataset `fraud_signals` is in scope, and the target must retain every valid late-arriving event. Which bounded tradeoff would preserve the core outcome and control operational risk?
- Maintain hourly full snapshots without primary keys
- Use SQL COPY INTO on the CRM REST URL directly
- Choose a managed source integration over custom polling while respecting connector limitations
- Use Auto Loader to monitor a folder that the CRM never exports into
Explanation
Correct Answer
C:Choose a managed source integration over custom polling while respecting connector limitations. This explicitly bounds the compromise while still addressing custom snapshots add extraction maintenance and cannot reliably detect late deletions. Acceptance relies on change, delete and re-create a source entity and verify incremental target convergence rather than simply declaring the change complete.
Incorrect Answers
A:Maintain hourly full snapshots without primary keys. Repeated full snapshots do not guarantee an accurate current-state target or deleted records. Monitoring after deployment does not make this approach a safe bounded tradeoff for `fraud_signals`.
B:Use SQL COPY INTO on the CRM REST URL directly. COPY INTO accepts supported file locations, not arbitrary CRM API requests. Monitoring after deployment does not make this approach a safe bounded tradeoff for `fraud_signals`.
D:Use Auto Loader to monitor a folder that the CRM never exports into. A file-ingestion service cannot read data that is not delivered as files. Monitoring after deployment does not make this approach a safe bounded tradeoff for `fraud_signals`.
Question 160
At a subscription media service, a supported CRM application offers a managed incremental connector, but the team plans to export full JSON snapshots hourly. The dataset `fraud_signals` is in scope, and the target must retain every valid late-arriving event. Before release, which acceptance test would actually demonstrate that the root problem is controlled?
- Review a platform-level availability report for `fraud_signals` without testing whether custom snapshots add extraction maintenance and cannot reliably detect late deletions.
- Change, delete and re-create a source entity and verify incremental target convergence
- Check that the intended configuration exists for `fraud_signals` and that a healthy run succeeded, without inducing the earlier failure.
- Run only a normal, unchanged `fraud_signals` sample and review CRM connector support matrix.
Explanation
Correct Answer
B:Change, delete and re-create a source entity and verify incremental target convergence. This reproduces the scenario-dependent condition, verifies CRM connector support matrix, update/delete capture, replication delay, and target governance, and directly tests whether custom snapshots add extraction maintenance and cannot reliably detect late deletions is resolved. Observed results, not intent, determine approval.
Incorrect Answers
A:Review a platform-level availability report for `fraud_signals` without testing whether custom snapshots add extraction maintenance and cannot reliably detect late deletions. Aggregate availability does not establish the behavior of affected `fraud_signals` records. It cannot replace the direct exercise: change, delete and re-create a source entity and verify incremental target convergence.
C:Check that the intended configuration exists for `fraud_signals` 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 custom snapshots add extraction maintenance and cannot reliably detect late deletions.
D:Run only a normal, unchanged `fraud_signals` sample and review CRM connector support matrix. A normal sample does not challenge custom snapshots add extraction maintenance and cannot reliably detect late deletions.