Microsoft DP-700 Practice Test 37

Question 360

Operations must compare all nightly Dataflow and Copy jobs across several workspaces. Select TWO sensible monitoring steps.

Use the Fabric monitoring hub or supported centralized workspace monitoring view and filter by workspace, item and run window.
Drill into the specific failed item’s activity/run details before concluding the cause.
Export the first workspace's recent run list and label it tenant-wide health.
Compare only Dataflow Gen2 refreshes while omitting Copy activity results.
Aggregate run counts without identifying the originating workspace or job identity.

Explanation

Correct Answers

Central run history provides consistent status and timing across monitored Fabric activities.

Aggregated status identifies candidate incidents, but root-cause evidence resides in item- and activity-level diagnostics.

Incorrect Answers

That excludes Dataflows and Copy jobs in all other workspaces and cannot prove global coverage.

The requested nightly review spans both workload types; omitting Copy loses part of the scheduled roster.

Anonymous counts can hide absent workspaces and duplicated runs, undermining completeness.

Question 361

A team checks one workspace’s recent runs and declares there were no failed nightly jobs across the tenant. What is the evidence gap?

The search scope excludes activities in other workspaces, so the absence of local failures cannot support a cross-workspace conclusion.
Success in one workspace proves all other Fabric workspaces have no failures.
Dataflow and Copy runs share a single global run ID across every workspace.
A single Dataflow's preview can represent the completion status of every nightly Copy job.

Explanation

Correct Answer

Monitoring conclusions must cover the actual inventory of participating workspaces and time windows.

Incorrect Answers

Workspace-local monitoring excludes jobs in other workspaces, so the tenant-wide conclusion is unsupported.

Run identifiers are execution-specific; coverage requires deliberate per-workspace aggregation or supported monitoring scope.

A preview is not a tenant-wide executed-run history and cannot attest to separate Copy jobs.

Question 362

A cross-workspace nightly health review is implemented. Which check validates report completeness?

Count successful runs in the first workspace but do not check jobs absent from the report.
Validate only the report's dashboard refresh timestamp.
Compare the monitoring results with the expected roster of scheduled items from each workspace and inspect a known controlled failure.
Check an arbitrary subset of workspaces without documenting the expected job roster.

Explanation

Correct Answer

Inventory reconciliation and an injected failure show the report is neither silently omitting workspaces nor filtering out failed run statuses.

Incorrect Answers

A partial roster can show no failures while omitting untracked scheduled jobs elsewhere.

Report freshness does not establish that the underlying inventory includes all expected workspaces and runs.

Completeness requires reconciling results against the authorized full set of nightly jobs.

Question 363

Multiple Fabric refreshes slow down at the same time that capacity consumption peaks. What evidence should the engineer correlate?

Inspect one notebook's Spark execution plan and ignore shared capacity concurrency.
Raise every activity timeout before measuring capacity utilization and throttling.
Conclude that unchanged input volumes prove there is no capacity-related slowdown.
Align refresh run timestamps with capacity metrics, throttling indicators and per-item resource demand before changing schedules or SKUs.

Explanation

Correct Answer

Time-correlated capacity telemetry distinguishes shared resource contention from a single faulty transformation.

Incorrect Answers

Several independent workloads slowing together suggests a shared resource effect beyond one notebook's plan.

Timeout extensions can hide but do not resolve contention at the observed peak.

Throughput can fall under concurrent saturation even with identical source data sizes.

Question 364

Several independent workloads slow simultaneously during a capacity saturation period. Their input data is unchanged. What is a likely shared cause?

Fabric capacity pressure or throttling is increasing queue/service time across concurrent workloads.
A change in a shared Fabric domain label simultaneously slows independent compute jobs.
The simultaneous slowdowns must be caused by identical source row growth in every workload.
A single notebook's Python library version necessarily controls every Dataflow refresh time.

Explanation

Correct Answer

Simultaneous delays across independent items are consistent with contention and should be confirmed against metered capacity metrics.

Incorrect Answers

Domain labels are governance metadata and do not consume the shared processing capacity as jobs execute.

Input volumes are unchanged; common capacity saturation is a more plausible shared bottleneck.

Independent workload engines can contend for the same capacity without sharing notebook package versions.

Question 365

A team reschedules refreshes to reduce peak contention. What is a meaningful before/after validation?

Compare average refresh latency only during an idle hour after rescheduling.
Compare capacity usage, overlap of scheduled runs, throttling frequency and median refresh duration across similar workload periods.
Measure total capacity usage but ignore retry, throttling and end-to-end runtimes.
Conclude success from one isolated fast refresh while other overlapping jobs remain unmeasured.

Explanation

Correct Answer

Controlled operational metrics test whether the schedule change reduced contention rather than merely hiding one long run.

Incorrect Answers

An idle-hour test doesn't demonstrate reduced contention at the former peak traffic interval.

Capacity graphs alone cannot establish that users experience fewer delays and missed SLAs.

The design addresses concurrent contention; validate the set of affected jobs and load windows.

Question 366

A Dataflow Gen2 refresh fails and the investigator must identify the exact source connection and query step responsible. What should they inspect?

Check only the Dataflow's destination columns after a rotated source credential.
Run an unrelated Dataflow preview as workspace Admin and infer the failed connection is repaired.
Open the failing refresh’s table/activity details and connection configuration, using available Mashup diagnostic logs to trace the connector error.
Replace the entire destination table before reading the per-query refresh error trace.

Explanation

Correct Answer

Refresh diagnostics include query and destination stages; the source connection must be identified from that run rather than guessed from item names.

Incorrect Answers

The failing source connection/gateway may reject authentication before destination writes begin.

A privileged alternate connection doesn't reproduce the scheduled identity and original source path.

Destructive sink changes won't identify the failed source connector and risk data loss.

Question 367

A dataflow fails after an endpoint credential rotates, but its author reviews only destination schema. What part of the failure chain is overlooked?

A destination table schema mismatch is the only possible cause of a Dataflow refresh failure.
The Dataflow query step can read a source without a valid connection because preview worked before.
The workspace needs a KQL query acceleration policy before it can authenticate a source.
The source connection identity/credential path is failing before data can reach the destination mapping stage.

Explanation

Correct Answer

Credential changes affect connector authentication and can stop Power Query evaluation before any sink write occurs.

Incorrect Answers

The error occurred after source credential rotation; source/gateway authentication must be examined.

Historical preview data doesn't attest to the current scheduled refresh identity or credentials.

KQL external-table acceleration is independent of Power Query source connection authorization.

Question 368

An incident report claims a wrong gateway connection caused a Dataflow Gen2 failure. Select TWO evidence checks that substantiate it.

Inspect the failed refresh logs for the actual connector endpoint, error and source query step.
Check that the destination schema is correct but never test the source connection.
Verify refresh success as a workspace Admin but skip the scheduled run identity.
Reproduce the source query using the corrected gateway/credential path and verify the intended dataflow refresh.
Compare only row counts before the original credential rotation, ignoring current source logs.

Explanation

Correct Answers

Run-scoped diagnostic evidence identifies which connection was attempted and where the failure occurred.

A successful controlled source read and refresh validate the identified root cause and fix.

Incorrect Answers

The incident is at the source/gateway authentication boundary, not necessarily at the output schema.

Admin access can mask the production credential or gateway problem and is not proof of scheduled recovery.

Historic counts cannot show that the repaired connection and current queries now execute successfully.

Question 369

A pipeline regularly exceeds its time budget, with Copy taking 5 minutes and a notebook transform taking 80 minutes. What should be analyzed first?

Spend effort reducing the five-minute Copy activity while leaving the 80-minute transform unchanged.
Inspect the notebook activity’s execution duration, stage diagnostics and its dependencies rather than tuning the already-fast Copy stage first.
Optimize unrelated Fabric dashboard refresh intervals as the primary SLA remedy.
Compare only individual Copy rates instead of pipeline-level activity duration and dependencies.

Explanation

Correct Answer

The longest critical-path activity is the primary contributor to elapsed time and should be measured before broad tuning.

Incorrect Answers

The transform dominates the critical path; further Copy tuning offers little end-to-end benefit.

Presentation cadence doesn't reduce the measured notebook transformation duration.

A healthy Copy step is not the dominant latency, so end-to-end profiling is necessary.

Question 370

A pipeline continues missing its SLA after Copy was optimized from five to three minutes. The transform still takes 80 minutes. Why is the gain small?

The optimized activity was not the dominant critical-path bottleneck; transformation runtime still controls nearly all end-to-end duration.
The pipeline's total time is dominated by the Copy step because it runs first.
Reducing Copy from five to three minutes should remove almost all total latency.
The remaining duration is a dashboard refresh display artifact rather than actual processing.

Explanation

Correct Answer

Reducing a minor component cannot materially improve the full run while the principal long-running activity remains unchanged.

Incorrect Answers

Copy is only five minutes; the downstream 80-minute transform dominates elapsed time.

Saving two minutes cannot substantially improve a pipeline still dominated by the long transformation.

Observed transform duration is 80 minutes, so instrument the activity instead of blaming presentation.

Implementing Data Engineering Solutions Using Microsoft Fabric (DP-700) Practice Tests

Microsoft DP-700 Practice Test 35 - 10 Questions (Last Reviewed 30-Sep-2026)
Microsoft DP-700 Practice Test 36 - 10 Questions (Last Reviewed 29-Sep-2026)
Microsoft DP-700 Practice Test 37 - 10 Questions (Last Reviewed 28-Sep-2026)
Microsoft DP-700 Practice Test 38 - 10 Questions (Last Reviewed 27-Sep-2026)
Microsoft DP-700 Practice Test 39 - 10 Questions (Last Reviewed 26-Sep-2026)

Pass Your Microsoft DP-700 Exam Easy!

Question 360

Explanation

Correct Answers

Incorrect Answers

Question 361

Explanation

Correct Answer

Incorrect Answers

Question 362

Explanation

Correct Answer

Incorrect Answers

Question 363

Explanation

Correct Answer

Incorrect Answers

Question 364

Explanation

Correct Answer

Incorrect Answers

Question 365

Explanation

Correct Answer

Incorrect Answers

Question 366

Explanation

Correct Answer

Incorrect Answers

Question 367

Explanation

Correct Answer

Incorrect Answers

Question 368

Explanation

Correct Answers

Incorrect Answers

Question 369

Explanation

Correct Answer

Incorrect Answers

Question 370

Explanation

Correct Answer

Incorrect Answers

Implementing Data Engineering Solutions Using Microsoft Fabric (DP-700) Practice Tests