Microsoft DP-700 Practice Test 33

Question 320

A joined Eventstream places events from different producers in inconsistent windows even though their physical times match. What is the likely cause?

The producers encode all events as plain local-time strings that can be directly compared lexically.
Source timestamps were parsed with inconsistent data types or time-zone assumptions before aggregation.
The event-time inconsistency arises because the database name is not UTC.
A later ingestion timestamp should always replace a supplied event timestamp for window attribution.

Explanation

Correct Answer

If one value is treated as local without its offset, the shared window sees a different instant than the other source.

Incorrect Answers

Equal physical instants with different offsets need normalization; lexical text order can differ from time order.

KQL database names have no authority over timestamp offset parsing or conversion semantics.

Ingestion time can differ from physical event time and cannot preserve the intended source-time windows.

Question 321

A multi-source timestamp normalization step is introduced. Select THREE verification checks.

Confirm both source timestamp strings parse into the expected supported datetime type.
Verify only that date-time parsing returns a nonnull value for each source.
Confirm equivalent instants with different offsets normalize to the same UTC time.
Compare local wall-clock hour strings without converting offsets to UTC.
Check one UTC event away from a date boundary and assume all source formats match.
Confirm near-midnight events enter the intended aggregate window after normalization.

Explanation

Correct Answers

Type consistency is required before time arithmetic and binning can be compared.

This detects the common error of stripping offset information without converting the instant.

Boundary events validate that the normalized event time, not the original display clock, controls the bucket.

Incorrect Answers

Successful parsing doesn't prove offset normalization or same-instant alignment across producers.

Wall-clock strings can differ for the same instant and misplace events near UTC midnight.

A single zero-offset event cannot expose incorrect nonzero-offset handling or boundary binning.

Question 322

A KQL query needs to enrich recent native alarm events with a small static device dimension in OneLake. Select TWO suitable design checks.

Cross join every alarm with all device-reference rows, then group by alarm ID.
Join alarms to device dimensions using only the nearest five-minute time bin.
Grant all KQL users database Admin to hide reference authorization errors.
Use a supported KQL join or accessible external-table reference to match device keys with dimension records.
Confirm the reference source authorization and benchmark join cardinality before enabling the production query.

Explanation

Correct Answers

KQL can combine hot Eventhouse data with compatible reference tables if both are queryable and appropriately keyed.

Wrong source access or a nonunique dimension key can cause query errors or multiply event rows.

Incorrect Answers

A cross join multiplies alarms, damaging cardinality and reference attribution.

Time proximity doesn't identify the correct device; use device keys and a unique effective reference row.

Broad privileges are unnecessary and do not correct duplicate reference keys or join semantics.

Question 323

A KQL join between alarm events and reference devices duplicates alarms. The dimension contains several rows per device key. What explains it?

The KQL join is duplicating alarms because the Eventhouse capacity is too small.
The join should use only a five-minute time bin instead of device identity.
The query lacks a Fabric domain administrator for each reference device.
The reference side is not unique at the join key, causing each event to match multiple dimension rows.

Explanation

Correct Answer

A one-to-many join expands the number of result rows; constrain or deduplicate reference versions first.

Incorrect Answers

Capacity limits may slow queries but do not create multiple dimension rows for one join key.

Time bins do not establish unique device lookup identity, so they cannot prevent repeated matches.

Governance roles do not remove duplicate dimension keys or fix join cardinality.

Question 324

A device enrichment query is ready for production. Which validation proves both correctness and scope?

Compare a fixed set of alarm IDs before and after the join, checking matched device attributes, row cardinality and behavior for unknown device IDs.
Compare only the total alarm count before enrichment and ignore repeated IDs afterwards.
Check that the device reference table contains records without checking key uniqueness.
Verify only that a KQL query executes with database Admin privileges.

Explanation

Correct Answer

Known matched and unmatched fixtures expose accidental many-to-many inflation and inappropriate record loss.

Incorrect Answers

Input count alone cannot prove the join preserved one row per alarm.

Record existence doesn't establish a one-to-one lookup or correct effective-version choice.

Execution and broad privileges don't establish proper device enrichment or least-privilege source access.

Question 325

A factory event feed must reach Eventhouse for queries and a separate lakehouse for downstream batch analysis. How should one Fabric Eventstream be configured?

Create a second producer stream and assign new event IDs for lakehouse-only records.
Create supported Eventstream branches or destinations to the Eventhouse and lakehouse, with each destination’s valid schema and permissions.
Add a lakehouse destination in the editor but leave the updated topology unpublished.
Repurpose the existing Eventhouse destination as lakehouse-only.

Explanation

Correct Answer

One source can route to multiple supported destinations without requiring the producer to publish duplicate messages.

Incorrect Answers

The goal is to branch one source with the same logical event IDs to two supported destinations.

Staged routing doesn't deliver live messages to the new sink until valid configuration is active.

Removing the Eventhouse branch would violate the requirement to deliver to both destinations.

Question 326

A producer sends one event, but only Eventhouse receives it even though a lakehouse destination was planned. The lakehouse branch was never attached/published. Why?

The Eventstream source must send the second destination's event with a new random event ID.
The missing lakehouse record proves Eventhouse queries are disabled in the workspace.
The active published stream topology has a destination path only to Eventhouse, so no event flow reaches the intended lakehouse sink.
The source is a different Fabric domain from the lakehouse and therefore cannot branch.

Explanation

Correct Answer

An unconnected or unpublished design branch is not part of the running event-routing graph.

Incorrect Answers

Changing logical IDs is unnecessary and undermines cross-sink record correlation for the same source event.

The Eventhouse branch is succeeding; the omitted lakehouse branch is a routing/publication issue.

Domain membership does not automatically create or prohibit the intended published destination connection.

Question 327

A second Eventstream destination is added. What validates branch delivery end to end?

Verify only that the event appears in Eventhouse, ignoring the lakehouse sink.
Check the Eventstream editor graph without publishing the changed topology.
Compare the two target item names but never reconcile source event IDs.
Emit one uniquely identified event, then locate the record in both intended destinations and check each destination’s published routing configuration.

Explanation

Correct Answer

A cross-destination event ID ties both outputs to the same producer message, proving independent sinks receive it.

Incorrect Answers

One successful branch cannot prove the newly added destination receives the same event.

A staged connection is not proof that the active stream routes messages to both sinks.

Matching item labels cannot demonstrate delivery or per-branch mapping of a specific message.

Question 328

A telemetry source delivers valid JSON mixed with malformed strings. The business needs valid records ingested and invalid payloads tracked. Which design is appropriate?

Implement supported parsing/type validation with explicit rejection, quarantine or error monitoring for malformed records before trusted storage.
Ingest every malformed string by disabling all field parsing checks.
Drop the entire stream whenever one input record contains malformed JSON.
Replace all invalid payloads with fabricated valid sensor values without diagnostic accounting.

Explanation

Correct Answer

Invalid JSON cannot safely be treated as a complete structured event; separating failures preserves valid traffic and diagnostic evidence.

Incorrect Answers

Invalid JSON can corrupt trusted schema assumptions; rejected records must be observed and handled.

This loses healthy events and fails the requirement to preserve valid records.

Invented data hides producer defects and prevents audit of malformed records.

Question 329

An Eventstream destination rejects an entire batch after encountering malformed JSON. There is no error-handling branch or parser rule. What explains the failure?

The batch failed because a single invalid JSON object automatically changes the warehouse capacity size.
The ingest mapping requires a valid structured payload and the pipeline has no policy for records that fail parsing.
The source record requires a Power BI semantic model before JSON can be parsed.
The error can be fixed by trusting all strings as valid typed integers without validation.

Explanation

Correct Answer

Schema parsing fails when required JSON syntax or fields are invalid; silent acceptance is not a reliable transformation.

Incorrect Answers

Malformed JSON is a parsing/mapping problem; capacity size cannot make an invalid payload valid.

A semantic model is unrelated to validating source syntax for Eventstream/Eventhouse ingestion.

Bypassing schema checks doesn't ensure meaningful typed fields and hides bad records rather than handling them.

Question 330

A bad-event handling path is added to Eventstream. Which fixture validates it?

Send only valid JSON examples and conclude error handling is correct.
Compare the Eventstream name before and after deploying parser rules.
Send one valid JSON event and one deliberately malformed payload; confirm the valid record arrives and the bad record is rejected or quarantined with observable diagnostics.
Drop every message and check that zero parser errors remain.

Explanation

Correct Answer

A mixed-validity test proves the pipeline isolates malformed messages without dropping or misinterpreting all traffic.

Incorrect Answers

Without malformed input the quarantine/reject branch has never been exercised.

Name continuity provides no evidence that bad payloads are isolated and observable.

Deleting good data alongside bad data is not successful validation of selective error handling.

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

Microsoft DP-700 Practice Test 31 - 10 Questions (Last Reviewed 04-Oct-2026)
Microsoft DP-700 Practice Test 32 - 10 Questions (Last Reviewed 03-Oct-2026)
Microsoft DP-700 Practice Test 33 - 10 Questions (Last Reviewed 02-Oct-2026)
Microsoft DP-700 Practice Test 34 - 10 Questions (Last Reviewed 01-Oct-2026)
Microsoft DP-700 Practice Test 35 - 10 Questions (Last Reviewed 30-Sep-2026)

Pass Your Microsoft DP-700 Exam Easy!

Question 320

Explanation

Correct Answer

Incorrect Answers

Question 321

Explanation

Correct Answers

Incorrect Answers

Question 322

Explanation

Correct Answers

Incorrect Answers

Question 323

Explanation

Correct Answer

Incorrect Answers

Question 324

Explanation

Correct Answer

Incorrect Answers

Question 325

Explanation

Correct Answer

Incorrect Answers

Question 326

Explanation

Correct Answer

Incorrect Answers

Question 327

Explanation

Correct Answer

Incorrect Answers

Question 328

Explanation

Correct Answer

Incorrect Answers

Question 329

Explanation

Correct Answer

Incorrect Answers

Question 330

Explanation

Correct Answer

Incorrect Answers

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