Microsoft Fabric Data Engineer Associate

Question 1

A data engineering team wants to migrate their existing Directed Acyclic Graphs (DAGs) written in Python into Microsoft Fabric to orchestrate a complex multi-stage ETL process that includes calls to external REST APIs and dependencies across multiple Fabric items. They require native support for their current DAG definitions with minimal rewriting. Which Fabric capability should they configure to meet this requirement?

Accepted Answer

Create an Apache Airflow job in the workspace and upload the existing DAG files

Answer

Create a Data pipeline and use the Web activity to replicate each DAG task manually

Answer

Configure a Dataflow Gen2 with sequential query dependencies to mimic the DAG structure

Answer

Use a notebook with the Fabric REST API and mssparkutils to chain notebook executions

Question 2

A governance team at Contoso must produce a monthly report showing which users viewed or downloaded a specific certified Power BI report hosted in a Fabric workspace, including exact timestamps and IP addresses. The Fabric Data Engineer needs to identify the correct source of this activity data and the role required to retrieve it. Which approach should be used?

Accepted Answer

Query the Microsoft Purview / Microsoft 365 unified audit log, which requires an Exchange Online audit role such as View-Only Audit Logs assigned to the retrieving account

Answer

Enable the workspace-level 'Usage metrics' report, which any workspace Member can open to see per-user view events with IP addresses

Answer

Assign the retrieving account the Fabric Admin role and read the OneLake diagnostic logs stored in the workspace's default lakehouse

Answer

Turn on Git integration for the workspace so that all report view and download events are committed to the connected repository

Question 3

A Fabric administrator notices that interactive report queries in a workspace have started returning delayed responses in the afternoon, and some background semantic model refreshes are being rejected. The workspace is assigned to an F64 capacity. The admin opens the Microsoft Fabric Capacity Metrics app to investigate. Which observation would most directly confirm that the capacity is throttling operations due to accumulated overutilization?

Accepted Answer

The 'Throttling' chart on the Compute page shows smoothed usage exceeding the interactive-delay and rejection limits, and the 'System events' table lists throttling events.

Answer

The number of active users in the workspace exceeds the per-capacity user license limit for F64.

Answer

The OneLake storage consumption for the workspace has exceeded the included storage quota for the SKU.

Answer

The 'Items' table shows several semantic models larger than the maximum model size supported by the SKU.

Question 4

Users of a Power BI report backed by a Fabric warehouse complain that visuals take an unusually long time to render during business hours. You open the Microsoft Fabric Capacity Metrics app and see that during peak periods the capacity shows sustained overloading, with the timepoint detail indicating a high volume of carryforward from earlier background operations. Interactive operations are experiencing an added delay before execution. Which throttling stage is the capacity currently in, and what does it mean for these interactive queries?

Accepted Answer

Interactive Delay — future interactive requests are delayed at submission time because recent overuse exceeds the 10-minute threshold, but they still complete.

Answer

Interactive Rejection — all interactive requests are immediately blocked and return errors until the overload clears.

Answer

Background Rejection — scheduled background jobs are rejected while interactive queries continue to run normally without any delay.

Answer

No throttling occurs; the delay is caused only by missing statistics on the warehouse tables and is unrelated to capacity utilization.

Question 5

A data engineering team needs to ingest CSV files from an Azure Data Lake Gen2 source, apply complex multi-step transformations using PySpark logic that reuses existing custom Python libraries, and write the results to a Lakehouse Delta table. The process must run every night at 2:00 AM and the team wants to minimize the number of components in the solution. Which Fabric item should they use to author and orchestrate this workload?

Accepted Answer

A notebook scheduled directly, or invoked by a pipeline on a schedule

Answer

A Dataflow Gen2 with a scheduled refresh

Answer

A pipeline using only the Copy data activity with a schedule trigger

Answer

An Apache Airflow job calling a stored procedure

Question 6

A data engineering team is building a new analytics solution in Microsoft Fabric. The bronze and silver layers require heavy transformation of large volumes of semi-structured JSON and Parquet files using PySpark. The gold layer will serve a set of downstream analysts and BI developers who are only comfortable writing T-SQL and require full multi-table INSERT/UPDATE/DELETE transactional support with multi-statement (ACID) transactions. The team wants to minimize the number of engines they manage while satisfying both groups. Which storage design should they choose?

Accepted Answer

Use a Lakehouse for the bronze and silver layers with Spark, then load the gold layer into a Warehouse for full T-SQL DML support.

Answer

Use a Lakehouse for all layers and expose the gold layer through the Lakehouse SQL analytics endpoint for the analysts.

Answer

Use a Warehouse for all layers and run the PySpark transformations against the Warehouse using the Spark connector.

Answer

Use a KQL database for the bronze and silver layers and a Lakehouse for the gold layer.

Question 7

A retail company has a Fabric lakehouse containing raw sales files in deeply nested JSON. The data engineering team must flatten and enrich this data on a nightly schedule. The transformation logic is complex, involving multiple joins, custom Python functions for currency conversion, and reuse of libraries already published to a Spark environment. The team consists of experienced Python developers who need full control over the transformation code and expect data volumes to grow to terabytes. Which transformation approach best fits these requirements?

Accepted Answer

Author a PySpark notebook that uses the shared Spark environment and schedule it in a pipeline

Answer

Build the transformation in a Dataflow Gen2 using the Power Query M interface

Answer

Load the JSON into a warehouse and use T-SQL OPENJSON with stored procedures

Answer

Ingest the JSON into a KQL database and transform with an update policy

Question 8

A Fabric warehouse contains a table named dbo.Employees that includes a Salary column. The HR analytics team must be able to query all other columns in the table, but they must be denied access to the Salary column specifically. You want to enforce this at the object level using T-SQL without creating a separate view or masking the data. What should you do?

Accepted Answer

Use DENY SELECT ON dbo.Employees(Salary) TO the HR analytics role while granting SELECT on the table.

Answer

Apply dynamic data masking to the Salary column for the HR analytics team.

Answer

Create a row-level security predicate function that filters out the Salary column.

Answer

Assign the HR analytics team the Viewer role on the workspace to hide the Salary column.

Question 9

A data engineer must load 200 large CSV files from an Azure Data Lake Storage Gen2 container into a Fabric Lakehouse Delta table on a nightly schedule. The files require no transformation—they must be landed exactly as-is into the table with the highest possible throughput and lowest compute overhead. Which approach should the engineer choose within a Fabric data pipeline?

Accepted Answer

Use a Copy activity in the pipeline to move the files directly into the Lakehouse table

Answer

Use a Dataflow Gen2 with staging enabled to read and write each file to the table

Answer

Use a Notebook activity running PySpark to read each CSV individually and append rows

Answer

Use a KQL query in an Eventhouse to ingest the CSV files into the Lakehouse table

Question 10

A data engineering team uses a Dataflow Gen2 in Fabric to ingest a large sales transaction table into a lakehouse. The source table grows by roughly 500,000 rows daily, but historical rows are never modified. Each scheduled refresh currently reprocesses the entire table, causing long run times and high capacity consumption. The team wants to reduce refresh duration while keeping the destination data accurate. Which approach should they configure?

Accepted Answer

Enable incremental refresh on the Dataflow Gen2 query using a DateTime column, defining a bucket range and a filter to only load newly arrived data

Answer

Disable staging on all queries in the Dataflow Gen2 so data is written directly to the destination

Answer

Convert the Dataflow Gen2 to a Copy activity in a pipeline with the Overwrite write behavior

Answer

Increase the Spark pool node size in the workspace Spark settings to speed up the refresh

Question 11

A Dataflow Gen2 that loads customer data into a lakehouse has started failing intermittently. The data engineer needs to determine the specific query and error message that caused the most recent failed refresh so it can be fixed. Which action provides the fastest, most detailed root-cause information for the failed run?

Accepted Answer

Open the Dataflow Gen2 refresh history and view the detailed error for the failed refresh, including the failing query and error message

Answer

Enable diagnostic logging on the destination lakehouse and inspect the Delta transaction log

Answer

Review the Capacity Metrics app to check whether the capacity was throttled during the refresh window

Answer

Re-run the entire dataflow with staging disabled and watch the applied-steps preview for errors

Question 12

A retail analytics team needs to combine a monthly Excel workbook, a SharePoint-hosted CSV, and a small on-premises SQL Server table into a single cleansed table in a Fabric Lakehouse. The transformation involves column renaming, type conversion, and merging the three sources. The team consists of business analysts who are comfortable with Power Query but have no experience writing PySpark or SQL code, and the data volumes are small (under 100,000 rows total). Which ingestion and transformation approach best fits this team and workload?

Accepted Answer

Author a Dataflow Gen2 that connects to all three sources, applies the transformations in the Power Query editor, and sets the Lakehouse table as the output destination

Answer

Create a Spark notebook that reads each source with the appropriate connector and merges them using PySpark DataFrame joins

Answer

Write a T-SQL stored procedure in a Fabric Warehouse to join and cleanse the sources, then copy the result to the Lakehouse

Answer

Build a KQL query in an Eventhouse to ingest and transform the three sources before landing them in the Lakehouse

Question 13

A data engineering team ingests raw CSV files into a lakehouse, applies complex PySpark transformations that reuse a shared library, and then loads curated data into a warehouse. Business analysts also need a low-code way to reshape a small reference dataset from a SharePoint list before the Spark job runs. The team wants a single orchestration that: runs the reference reshaping first, then the heavy transformation, then the warehouse load, with each step only starting after the previous one succeeds. Which combination best fits these requirements?

Accepted Answer

Create a pipeline that calls a Dataflow Gen2 for the SharePoint reference reshaping, then a notebook for the PySpark transformations, then a Copy/warehouse load activity, chained with success dependencies.

Answer

Build one Dataflow Gen2 that performs the SharePoint reshaping, the PySpark transformations, and the warehouse load in sequence.

Answer

Schedule three independent notebooks with staggered start times so they run in the correct order.

Answer

Use a single notebook that reads the SharePoint list, runs the Spark transformations, and writes to the warehouse, triggered on file arrival.

Question 14

A data engineer is preparing data in a Fabric lakehouse to feed a star-schema dimensional model. The source is a set of highly normalized operational tables (Customers, Addresses, Regions, and SalesReps) that must be combined into a single wide Customer dimension table with columns from all four source tables. The transformation involves multiple joins and needs to run nightly over hundreds of millions of rows. Which approach best prepares this denormalized dimension for the dimensional model?

Accepted Answer

Create a PySpark notebook that joins the four source tables and writes the denormalized result to a Delta table in the Gold layer

Answer

Create a OneLake shortcut to each of the four source tables so the dimensional model can reference them directly without physical joins

Answer

Use a KQL query in an Eventhouse to aggregate the four tables into a materialized view for the dimension

Answer

Keep the four normalized tables as-is and build the join logic inside the semantic model's relationships at query time

Question 15

Your team uses a Fabric deployment pipeline with three stages: Development, Test, and Production. A Development-stage workspace contains a lakehouse, a warehouse, and a semantic model that connects to the warehouse. After you deploy all items from Development to Test for the first time, you notice the semantic model in Test still points to the warehouse in the Development workspace instead of the one in Test. You want future and current deployments to automatically repoint the semantic model to the warehouse within the same target stage without manually editing connections after every deployment. What should you do?

Accepted Answer

Configure a data source deployment rule on the semantic model in the Test stage to point to the Test warehouse.

Answer

Rely on the deployment pipeline's automatic binding, which repoints dependencies when all related items are deployed together in the same operation.

Answer

Recreate the semantic model manually in the Test workspace and delete the deployed copy.

Answer

Enable Git integration on the Test workspace so the connection strings sync from the repository.

Question 16

A data engineering team uses a Fabric deployment pipeline with Development, Test, and Production stages. Before promoting changes from Test to Production, the release manager wants to see exactly which items differ between the two stages so only intended changes are deployed. Which action lets the release manager view the differences between Test and Production before deploying?

Accepted Answer

Use the comparison indicator shown between the Test and Production stages in the deployment pipeline, which flags items as different, new, or identical

Answer

Open the Git integration panel in the Production workspace and review uncommitted changes

Answer

Run a Test-to-Production deployment with 'compare only' mode, which validates without publishing

Answer

Export both workspaces to a database project and use schema compare to diff the items

Question 17

A data engineering team manages a Fabric workspace containing lakehouses, notebooks, and semantic models. They want to enforce peer review of every code change through pull requests before changes are merged, while also promoting validated content across Development, Test, and Production workspaces without manually rebuilding items. Which combination of lifecycle management features should they configure to satisfy BOTH requirements?

Accepted Answer

Use Git integration with a feature-branch workflow for pull-request review, and use deployment pipelines to promote items across the three workspace stages.

Answer

Use deployment pipelines alone, relying on stage comparison and deployment rules to handle code review and promotion.

Answer

Use Git integration alone, committing directly to the main branch and syncing each workspace to a different branch for promotion.

Answer

Export items as database projects, review the SQL in a repository, and manually import them into each workspace stage.

Question 18

Your team uses a Fabric deployment pipeline with Development, Test, and Production stages. A data engineer updated three notebooks and one semantic model in the Development stage, but a colleague warns that a fourth notebook still in Development is experimental and must NOT reach Test yet. Before deploying, you want to verify exactly which items differ between Development and Test and push only the approved changes. Which approach meets these requirements with the least risk?

Accepted Answer

Use the deployment pipeline's compare view to identify changed items, then deploy only the selected approved items to the Test stage.

Answer

Deploy the entire Development stage to Test, then manually delete the experimental notebook from the Test stage afterward.

Answer

Create a Git branch from Development containing only the approved items and merge it directly into the Test workspace.

Answer

Configure a deployment rule on the experimental notebook so it is skipped during the deployment to Test.

Question 19

A data engineering team maintains a Fabric warehouse whose schema (tables, views, stored procedures) changes frequently. They want developers to author schema changes locally in Visual Studio Code using SQL, review those changes through pull requests before merging, and validate the resulting schema against a target before publishing. Which lifecycle management approach best satisfies these requirements?

Accepted Answer

Use a Fabric database project (SQL project) integrated with Git, allowing developers to author and build schema changes locally and publish validated changes to the warehouse.

Answer

Use a Fabric deployment pipeline with three stages and rely on the automatic schema comparison to author changes directly in the production stage.

Answer

Connect the workspace to Git and have developers edit warehouse tables directly in the Fabric portal, committing changes after each edit.

Answer

Export the warehouse as a Dataflow Gen2 and version the query definitions in an external repository.

Question 20

Your team uses a Fabric deployment pipeline with three stages: Development, Test, and Production. A data pipeline in the workspace uses a copy activity that connects to a SQL source. In Development the connection points to a dev SQL server, but after deploying to Production the item must automatically point to the production SQL server without any manual edits after each deployment. What should you configure to ensure the correct server is always used in the Production stage?

Accepted Answer

Create a deployment rule on the Production stage that parameterizes the data pipeline's connection to the production SQL server.

Answer

Manually update the connection in the Production workspace after every deployment.

Answer

Enable a Git branch for the Production stage and change the server value in the pipeline JSON before each merge.

Answer

Add a sensitivity label to the Production data pipeline so it uses the correct server.

Question 21

A data engineering team uses Git integration to manage source control for their Fabric workspace, which is connected to a feature branch. Separately, the team lead has configured a deployment pipeline with Development, Test, and Production stages. A developer completes changes in the workspace, commits them to Git, and now needs to promote validated content from Development to Test. Which statement correctly describes how these two lifecycle management features work together in this scenario?

Accepted Answer

Git integration handles versioning and collaboration for a single workspace, while deployment pipelines promote content between separate workspaces assigned to each stage.

Answer

Deployment pipelines and Git integration are mutually exclusive; enabling one automatically disables the other for the workspace.

Answer

Committing changes to Git automatically triggers deployment to the next pipeline stage without any manual promotion step.

Answer

Deployment pipelines read content directly from the connected Git branch rather than from the workspace, so the commit must be merged to main before promotion.

Question 22

Your team uses a Fabric deployment pipeline with Development, Test, and Production stages. After finishing several updates in Development, a data engineer notices that only two of the five modified items are ready to promote to Test; the other three are still being validated. The engineer wants to move just the two finished items to Test without affecting the unvalidated items already in Test. What should the engineer do?

Accepted Answer

Use the deployment pipeline compare view to select only the two ready items and perform a selective deployment to the Test stage.

Answer

Deploy all changed items to Test, then manually delete the three unvalidated items from the Test stage afterward.

Answer

Create a separate deployment pipeline containing only the two items and deploy that pipeline to Test.

Answer

Commit only the two items to the Git-connected branch, then deploy the entire Development stage to Test.

Question 23

A data engineering team maintains a Fabric workspace containing a Data Factory pipeline that ingests raw files, a notebook that transforms them, and a semantic model refresh. The transformation notebook must run only after the ingestion pipeline completes successfully, and business users need the entire flow to run automatically every night at 2:00 AM. The team wants a single orchestration artifact that controls the end-to-end run order and can pass the ingestion date as a parameter into the notebook. What should the team implement?

Accepted Answer

Create a Data Factory pipeline that chains a Copy activity, a Notebook activity (with a base parameter for the ingestion date), and a Semantic model refresh activity, then attach a scheduled trigger for 2:00 AM.

Answer

Schedule the ingestion pipeline, the notebook, and the semantic model refresh independently, each with its own 2:00 AM schedule so they all start at the same time.

Answer

Use a Dataflow Gen2 to perform ingestion and transformation together, then rely on the semantic model's automatic refresh to complete the flow.

Answer

Add the notebook's transformation code as a cell inside the ingestion pipeline's Copy activity and enable Airflow to trigger it hourly.

Question 24

You manage a Fabric deployment pipeline with three stages: Development, Test, and Production. Each stage is assigned to a separate workspace. After a colleague made changes in the Development workspace, you open the deployment pipeline view and notice an orange 'X' comparison indicator between the Development and Test stages, while the Test-to-Production comparison shows a green checkmark. You need to promote only the reviewed changes and verify what will move. What does the orange indicator signify, and what is the correct next step?

Accepted Answer

There are differences between the Development and Test stages; select the changed items and deploy from Development to Test to synchronize them.

Answer

The stages are identical; you must unassign and reassign the Test workspace before any deployment can occur.

Answer

The Test workspace has unsupported items; you must delete those items in Test and then run a full backward deployment from Production.

Answer

The deployment pipeline is broken; you must recreate the pipeline and re-import all items from Git to clear the indicator.

Question 25

A data engineering team wants to formalize their release process for Fabric items. They need three separate environments: one for active development, one where QA testers validate changes before release, and one that serves production reports to business users. They plan to use a Fabric deployment pipeline to move content between these environments. What is the minimum configuration they should set up to meet this requirement?

Accepted Answer

Create a deployment pipeline with three stages (Development, Test, Production), each assigned to a separate workspace.

Answer

Create three separate deployment pipelines, one per environment, and manually copy items between them.

Answer

Create a single workspace and use Git branches to represent Development, Test, and Production.

Answer

Create a deployment pipeline with two stages and reuse the Test stage workspace for production reporting.

Question 26

You manage a Fabric Warehouse containing a customer table with an EmailAddress column. Compliance requires that most analysts see only masked email values, but a small group of fraud investigators must see the full unmasked values. You apply dynamic data masking to the EmailAddress column. After applying the mask, the fraud investigators report they still cannot see the real email addresses. What must you do so that the fraud investigators see unmasked data while other analysts continue to see masked values?

Accepted Answer

Grant the UNMASK permission to the fraud investigators' principals in the Warehouse.

Answer

Add the fraud investigators to the workspace Admin role so masking no longer applies to them.

Answer

Remove the mask from the column and instead create a column-level security rule that hides it from analysts.

Answer

Assign the fraud investigators the Viewer role on the workspace and re-run the masking definition.

Question 27

Your Fabric tenant has multiple domains, and a governance team wants to formally recognize high-quality datasets. A data engineer has built a semantic model that has passed all quality reviews and should be marked as 'Certified' so it appears as the authoritative source in the OneLake data hub. The engineer attempts to set the certification but only sees the 'Promoted' option available. What must happen for the model to be marked as Certified?

Accepted Answer

The engineer must be added to the list of authorized certifiers defined by a Fabric administrator in the tenant or domain settings.

Answer

The engineer must be assigned the Admin role on the workspace containing the semantic model.

Answer

The semantic model must first be endorsed as 'Promoted' before it becomes eligible to be certified.

Answer

A sensitivity label of 'Confidential' or higher must be applied to the semantic model before certification is allowed.

Question 28

A data engineering team receives sales files from partners at unpredictable times throughout the day. The files are uploaded to a specific folder in a OneLake lakehouse Files section. The team wants a Fabric Data pipeline to run automatically within moments of a new file arriving, without polling on a fixed schedule. Which approach should they configure to trigger the pipeline?

Accepted Answer

Create a Fabric event-based trigger (using Real-Time hub / Fabric events) that listens for OneLake events and starts the pipeline on file creation

Answer

Configure a scheduled trigger on the pipeline set to run every 5 minutes and add a Get Metadata activity to check for new files

Answer

Enable Git integration on the workspace so that commits trigger the pipeline automatically

Answer

Use a Dataflow Gen2 with incremental refresh, which automatically detects and processes new files as they land

Question 29

An Eventhouse in your Real-Time Intelligence solution stores 18 months of telemetry data for compliance, but analysts only run interactive dashboards against the most recent 30 days. Queries against recent data have become noticeably slower over time, and capacity monitoring shows high memory pressure on the Eventhouse. You need to improve interactive query performance for recent data while retaining all 18 months for occasional audits, keeping costs reasonable. What should you configure?

Accepted Answer

Set the hot cache (caching policy) to 30 days while keeping the retention policy at 18 months on the KQL database or table.

Answer

Reduce the retention policy to 30 days and export older data to a lakehouse via a scheduled pipeline.

Answer

Set both the hot cache policy and the retention policy to 18 months so all data is available for fast queries.

Answer

Create a materialized view that filters the last 30 days and query the view instead of the base table.

Question 30

You manage an Eventhouse (KQL database) that ingests IoT telemetry through a data connection. Overnight, downstream dashboards show no new rows for one table, while other tables continue to update normally. You need to quickly determine whether ingestion for that specific table is failing and see the underlying error messages. Which action should you take first?

Accepted Answer

Run the .show ingestion failures command against the affected table to review recent failed ingestion operations and their error reasons

Answer

Increase the hot cache retention period on the affected table to force the ingested data to reappear in queries

Answer

Drop and recreate the table so that the ingestion pipeline starts writing fresh data

Answer

Create a materialized view over the table to deduplicate the missing records

Question 31

An Eventhouse in Microsoft Fabric receives a continuous stream of small event batches from an Eventstream. Analysts complain that newly ingested data does not appear in query results for several minutes. On investigation, you find the KQL database table uses the default ingestion batching settings. You need to reduce the end-to-end latency between event arrival and query availability without significantly increasing ingestion cost per record. Which action should you take?

Accepted Answer

Lower the batching policy's time and size thresholds so batches are sealed and ingested sooner.

Answer

Increase the hot cache retention period on the table so recent data is cached longer.

Answer

Create a materialized view over the table to accelerate the analysts' queries.

Answer

Enable an update policy to transform the records as they are ingested.

Question 32

An Eventhouse ingests millions of IoT telemetry rows per hour into a KQL table. A dashboard runs a query that computes the count of events per device over the last 24 hours, and this same aggregation query is executed hundreds of times per minute by many concurrent users. Users report the dashboard is slow and consuming excessive query resources. You need to reduce query latency and resource consumption for this repeated aggregation with minimal ongoing maintenance. What should you do?

Accepted Answer

Create a materialized view that pre-aggregates the event count per device, and have the dashboard query the materialized view.

Answer

Increase the hot cache retention period on the source table so more raw data stays in memory.

Answer

Add an update policy that copies every ingested row into a second table for the dashboard to query.

Answer

Schedule a pipeline to export the aggregated results to a Lakehouse Delta table every hour.

Question 33

A retail company runs an Azure SQL Database that records order transactions. The data engineering team wants to continuously stream row-level inserts, updates, and deletes from this operational database into a Fabric Eventstream so that downstream analytics reflect changes with minimal latency, without periodically re-scanning the entire table. Which Eventstream source type should the team configure?

Accepted Answer

An Azure SQL Database change data capture (CDC) source

Answer

A sample data source that generates simulated order events

Answer

A OneLake shortcut pointing to the Azure SQL Database

Answer

An Azure Event Hubs source reading from the database transaction log

Question 34

A retail company ingests point-of-sale transactions into a Fabric Eventstream from an Azure Event Hubs source. The data engineering team must route the events so that transactions with an amount greater than 5000 are sent to a KQL database for real-time fraud analysis, while all transactions (regardless of amount) are landed into a lakehouse table for batch reporting. They want to accomplish this entirely within the Eventstream without writing Spark structured streaming code. What is the most appropriate way to implement this?

Accepted Answer

Add a Filter event processor to create a derived stream for high-value transactions routed to the KQL database (eventhouse) destination, and add a separate destination from the source stream to the lakehouse table.

Answer

Configure a single lakehouse destination and use a downstream Dataflow Gen2 to filter and copy high-value rows into the KQL database on a schedule.

Answer

Use a Spark structured streaming notebook to read from Event Hubs, apply the amount filter, and write to both the KQL database and lakehouse in foreachBatch.

Answer

Add an Aggregate event processor to group transactions by amount and route the aggregated output to both the KQL database and lakehouse destinations.

Question 35

A retail company streams clickstream events from an Azure Event Hub into Microsoft Fabric using an Eventstream. Analysts need to run sub-second, ad-hoc analytical queries over the last 90 days of raw events, including time-series aggregations and pattern-matching functions, with data available for query within seconds of arrival. Which Eventstream destination should you configure to best meet these requirements?

Accepted Answer

An Eventhouse (KQL Database) destination, querying with KQL

Answer

A Lakehouse destination writing Delta tables, queried with the SQL analytics endpoint

Answer

A Warehouse destination populated via a scheduled pipeline copy, queried with T-SQL

Answer

A Dataflow Gen2 destination that materializes events into a staging table

Question 36

A telemetry team ingests IoT sensor readings through an Eventstream into an Eventhouse. The analysts need a KQL query that reports the average temperature over the last 5 minutes, but the result must be recalculated and emitted every 1 minute so dashboards show a smooth, continuously updating rolling average. Which windowing approach should you configure?

Accepted Answer

A hopping window with a 5-minute size and a 1-minute hop

Answer

A tumbling window with a 5-minute size

Answer

A session window with a 5-minute timeout

Answer

A snapshot window that captures the latest value every minute

Question 37

A retail company streams point-of-sale transactions into a Fabric Eventstream. Analysts need to run sub-second KQL queries against the incoming data with full indexing and the lowest possible query latency, while also making the same data available in the lakehouse for periodic Spark batch reporting without duplicating storage. Which approach should you implement for the Eventhouse destination?

Accepted Answer

Land the data in a native Eventhouse (KQL) table for low-latency queries, then create a OneLake shortcut from the lakehouse to that table for Spark batch reporting

Answer

Create a OneLake shortcut in the Eventhouse pointing to a lakehouse Delta table, and query the shortcut with KQL

Answer

Write the Eventstream output only to a lakehouse Delta table and query it directly with KQL from the Eventhouse

Answer

Copy the data into both a native Eventhouse table and a separate lakehouse Delta table using a scheduled pipeline

Question 38

A retail analytics team ingests clickstream events into a Fabric Eventhouse using an Eventstream. Analysts run frequent, low-latency KQL queries that filter and aggregate the last 24 hours of events. The team wants the fastest possible query performance and full indexing/caching benefits on this hot data. When configuring the Eventstream destination into the Eventhouse, which approach should they choose?

Accepted Answer

Land the data as a native KQL table (direct ingestion) in the Eventhouse

Answer

Create a OneLake shortcut in the Eventhouse pointing to a lakehouse Delta table populated by the Eventstream

Answer

Write the events to a lakehouse table via Spark structured streaming, then query it from the Eventhouse over a shortcut

Answer

Store the events as Parquet files in OneLake and query them through an external table

Question 39

An Eventstream ingests telemetry from an Azure Event Hubs source and routes it to a KQL database destination. Over the past hour, the source is receiving events normally, but no new rows appear in the destination table. The upstream Event Hubs metrics show healthy throughput. Where should you look FIRST to identify the root cause of the delivery failure?

Accepted Answer

The Eventstream's runtime logs and the data insights on the destination node in the Eventstream editor

Answer

The capacity metrics app to check for interactive query throttling on the workspace

Answer

The semantic model refresh history for the connected report

Answer

The Azure Event Hubs consumer group partition offsets in the Azure portal

Question 40

A retail analytics team already lands IoT sensor telemetry into a Delta table in a Fabric lakehouse using Spark Structured Streaming. The KQL analysts now need to run interactive, low-latency queries over the same telemetry from an Eventhouse (KQL database) without physically duplicating the data or maintaining a second ingestion pipeline. Data volume is high and the analysts are comfortable with slightly higher query latency than a native KQL table would provide. Which approach best meets these requirements?

Accepted Answer

Create a OneLake shortcut in the KQL database that points to the existing lakehouse Delta table

Answer

Add a new Eventstream destination that writes the telemetry into a native KQL table in the Eventhouse

Answer

Use a Data pipeline Copy activity to periodically copy the Delta table into a native KQL table

Answer

Enable database mirroring on the lakehouse to replicate the Delta table into the Eventhouse

Question 41

A retail company streams point-of-sale transactions into a Microsoft Fabric Eventstream. The business wants only per-store total sales amounts, computed in 5-minute non-overlapping intervals, written to an Eventhouse for near real-time dashboards. Raw individual transactions must NOT be persisted to reduce storage. Which approach should you implement inside the Eventstream?

Accepted Answer

Add a Group by operator in the Eventstream event processor using a 5-minute tumbling window grouped by store, then route the aggregated output to the Eventhouse destination.

Answer

Route the raw stream directly to the Eventhouse destination and build a KQL update policy on the destination table to aggregate after ingestion.

Answer

Add a Group by operator in the Eventstream using a 5-minute hopping window with a 1-minute hop, then route the output to the Eventhouse.

Answer

Ingest the raw stream into a Lakehouse via a shortcut and use a scheduled Spark structured streaming job to aggregate every 5 minutes.

Question 42

A retail company ingests point-of-sale transactions into a Fabric Eventstream. The analytics team needs a per-store total sales figure computed for every fixed 5-minute interval, with no overlap between intervals. Due to intermittent network issues at some stores, a small percentage of events arrive up to 90 seconds after their event-time window has closed. The team wants these late events to still be included in the correct interval's aggregate rather than dropped. Which combination of windowing configuration should you use in the Eventstream aggregation?

Accepted Answer

A tumbling window of 5 minutes with a configured late-arrival tolerance (grace period) of 90 seconds

Answer

A hopping window with a 5-minute size and a 90-second hop, ignoring late-arrival settings

Answer

A sliding window of 5 minutes that recalculates on every incoming event

Answer

A session window with a 90-second timeout gap between events

Question 43

A logistics company needs a solution in Microsoft Fabric to process GPS pings from thousands of delivery trucks. The data arrives continuously, and dispatchers need a live dashboard showing each truck's most recent location with under 10 seconds of latency. The engineering team also wants to route the same incoming pings to a Lakehouse for later historical analysis. Which approach best satisfies both the low-latency dashboard and the historical archive requirements with the least custom code?

Accepted Answer

Create an Eventstream that ingests the GPS pings, then add both an Eventhouse (KQL Database) destination for real-time querying and a Lakehouse destination for historical storage.

Answer

Schedule a Dataflow Gen2 to run every minute that reads the GPS pings and writes them to a Lakehouse table used by the dashboard.

Answer

Use a Data pipeline with a Copy activity on a 5-minute schedule to land the pings in a Warehouse and build the dashboard on top of it.

Answer

Write a PySpark structured streaming job that reads the pings and writes only to a Lakehouse Delta table, then connect the dashboard directly to that table.

Question 44

A retail analytics team ingests clickstream events into an Eventhouse via an Eventstream. Raw JSON events land in a table named RawEvents. The team needs every incoming row to be automatically parsed, filtered to remove bot traffic, and written to a curated table with a flattened schema — with the transformation applied at ingestion time inside the KQL database, requiring no external orchestration and minimal added latency. Which approach should they use?

Accepted Answer

Create a KQL update policy on the curated table that runs a transformation query against RawEvents at ingestion time

Answer

Schedule a Dataflow Gen2 to read RawEvents every 5 minutes, transform, and load the curated table

Answer

Add a Spark Structured Streaming job that reads the Eventhouse table and writes the curated output

Answer

Create a KQL materialized view over RawEvents to deduplicate and flatten the data

Question 45

A retail analytics team ingests clickstream data that arrives continuously from a web application into an Azure Event Hub. Business users need dashboards refreshed with a maximum data latency of a few seconds, and the incoming events must be filtered and enriched before landing in an Eventhouse for interactive KQL queries. The team wants a low-code approach that avoids managing Spark clusters. Which ingestion and transformation approach best fits these requirements?

Accepted Answer

Configure a Fabric Eventstream to consume the Event Hub, apply filter and manage-fields transformations in the stream, and route the output to an Eventhouse destination

Answer

Use a Data Factory pipeline with a Copy activity on a 5-minute schedule to load events into a Lakehouse, then query with the SQL analytics endpoint

Answer

Create a Dataflow Gen2 with incremental refresh scheduled every minute to pull events from the Event Hub into a Warehouse

Answer

Write a Spark Structured Streaming notebook that reads from the Event Hub, applies transformations, and writes to a Lakehouse Delta table

Question 46

A retail company needs to process a continuous feed of point-of-sale transactions arriving through Azure Event Hubs. The requirement is to land the raw events, apply lightweight filtering and column mapping, and route the results to both a KQL database for real-time dashboards and a lakehouse Delta table for later batch analytics. The team wants a low-code solution that avoids writing and maintaining custom code, and they have no complex stateful transformation needs. Which Fabric feature should they use to implement this pipeline?

Accepted Answer

Eventstream with multiple destinations and event processing transformations

Answer

A Spark structured streaming notebook reading from Event Hubs and writing to both sinks

Answer

A Dataflow Gen2 with a scheduled refresh polling the Event Hub

Answer

A Data pipeline Copy activity with a tumbling window trigger

Question 47

A data engineering team ingests a small vendor reference table (about 5,000 rows) into a Fabric Lakehouse Delta table each night. The source system does not expose any reliable modified-date, version, or change-tracking metadata, and vendor records can be silently updated or deleted between runs. The team needs the Lakehouse table to always exactly match the current source state after each nightly run, while keeping the pipeline logic as simple as possible. Which loading pattern should they implement?

Accepted Answer

A full load that overwrites (truncate and reload) the entire Delta table on every run

Answer

An incremental load using a watermark on a surrogate row-insert timestamp added at ingestion

Answer

A MERGE-based upsert that matches on the business key and updates changed rows

Answer

An append-only load that inserts all source rows into the Delta table each run

Question 48

You manage a Fabric workspace that contains notebooks, a lakehouse, and semantic models. Your team wants to track all changes to these items using an existing Azure DevOps repository, allow developers to work in feature branches, and merge approved changes back into the workspace. Which action correctly establishes source control for this workspace?

Accepted Answer

Connect the workspace to the Azure DevOps repository through Workspace settings > Git integration, selecting the organization, project, repository, and branch.

Answer

Create a deployment pipeline with development, test, and production stages and assign the workspace to the development stage.

Answer

Export each item as a .pbip file and manually upload the files to the Azure DevOps repository.

Answer

Enable the OneLake catalog and share the workspace items with the DevOps service principal as a collaborator.

Question 49

A retail company ingests daily order files into a Fabric lakehouse Delta table. Due to an upstream system retry mechanism, some order records are delivered multiple times across different daily files, and occasionally an order file arrives a day late containing corrected versions of orders already loaded earlier. Each order has a unique OrderId and a LastModifiedTimestamp. The engineering team needs the Bronze-to-Silver PySpark transformation to keep only the most recent version of each order and eliminate duplicates. Which approach best meets this requirement?

Accepted Answer

Use a MERGE operation keyed on OrderId, updating the target row only when the source LastModifiedTimestamp is greater than the existing target timestamp.

Answer

Append all incoming records to the Silver table and rely on Delta's automatic deduplication during OPTIMIZE compaction.

Answer

Apply dropDuplicates on OrderId before writing, then overwrite the entire Silver table on each run.

Answer

Filter the source to include only records whose ingestion date equals the current processing date, then append them to the Silver table.

Question 50

A Fabric data engineer maintains a Lakehouse Delta table called Customers with 40 million rows. A daily pipeline extracts only rows that changed since the previous run (based on a LastModified column) into a staging Delta table. The engineer must apply these changes to the target Customers table so that existing customers are updated in place and brand-new customers are inserted, without rewriting the entire 40-million-row table each day. Which approach in a PySpark notebook best meets this requirement?

Accepted Answer

Use a Delta Lake MERGE operation matching on the customer key, with WHEN MATCHED THEN UPDATE and WHEN NOT MATCHED THEN INSERT clauses.

Answer

Overwrite the target Customers table with the union of the existing rows and the staging rows using DataFrame.write.mode('overwrite').

Answer

Append the staging rows to the target table using DataFrame.write.mode('append') and rely on Delta to deduplicate automatically.

Answer

Truncate the target Customers table and reload it entirely from the source system each day.

Question 51

A data engineer must incrementally load a large SQL Server 'Orders' table into a Fabric Lakehouse every hour. The source table has an 'OrderId' identity column and a 'LastModifiedDate' datetime column that is updated whenever an order row is inserted or updated. The pipeline must capture both newly inserted and previously updated rows since the last successful run, while minimizing the volume of data transferred. Which incremental loading strategy should the engineer implement?

Accepted Answer

Use a high-watermark pattern based on the 'LastModifiedDate' column, storing the maximum value from each run and filtering the next run for rows greater than that stored value.

Answer

Use a high-watermark pattern based only on the 'OrderId' identity column, storing the maximum OrderId and filtering for rows greater than it on the next run.

Answer

Perform a full load of the entire 'Orders' table each hour and overwrite the destination Delta table.

Answer

Enable a scheduled Dataflow Gen2 refresh that truncates and reloads the table without any filter predicate.

Question 52

A data engineering team receives a nightly drop of 2 TB of raw Parquet files in an Azure Data Lake Storage Gen2 account owned by a partner organization. The team must run heavy PySpark transformations (joins, aggregations, and deduplication) over this data every night, but they want to avoid physically copying the 2 TB into their own Fabric workspace and want to minimize storage duplication and ingestion latency. Which approach should they implement?

Accepted Answer

Create a OneLake shortcut in a lakehouse pointing to the ADLS Gen2 location, then run the PySpark transformations against the shortcut.

Answer

Use a Copy activity in a Data Factory pipeline to load the full 2 TB into a lakehouse Files section each night before transforming.

Answer

Configure database mirroring from the ADLS Gen2 account into a Fabric warehouse and transform with T-SQL.

Answer

Build a Dataflow Gen2 that imports the Parquet files with staging enabled and applies the transformations in Power Query.

Question 53

A data engineer is ingesting IoT telemetry into a KQL database in an Eventhouse. The source files in a OneLake shortcut are JSON documents where each file contains a single JSON object spanning multiple lines (pretty-printed), not one object per line. An initial ingestion attempt using the default settings produced parsing errors and malformed rows. Which ingestion format setting should the engineer use so the files are parsed correctly?

Accepted Answer

Set the ingestion format to 'multijson' so the parser treats the entire file content as a whole JSON document rather than line-delimited records

Answer

Set the ingestion format to 'json', which expects one JSON object per line (newline-delimited)

Answer

Set the ingestion format to 'csv' and rely on an ingestion mapping to split the JSON fields

Answer

Set the ingestion format to 'raw' and parse the JSON later with an update policy

Question 54

A telemetry pipeline lands IoT device readings into a KQL database table in an Eventhouse. Devices frequently resend the same reading (identified by DeviceId and ReadingTimestamp) due to network retries, producing duplicate rows. Analysts must always query only the most recent version of each reading with minimal query latency, and the deduplication logic should be precomputed and kept current automatically as new data arrives. What should you implement?

Accepted Answer

A materialized view using arg_max() over the ingestion time, grouped by DeviceId and ReadingTimestamp

Answer

An update policy that writes deduplicated rows into a second table on each ingestion batch

Answer

A stored function that applies summarize arg_max() at query time whenever analysts run reports

Answer

A retention policy that automatically deletes older duplicate rows after a short soft-delete window

Question 55

A retail company streams point-of-sale transactions into an Eventhouse (KQL database) in Microsoft Fabric. Analysts need a query that, for each store, continuously computes the total sales amount over the most recent 15 minutes, recalculated every 5 minutes, so overlapping periods are reported to detect sudden spikes. Which KQL windowing approach should you use to satisfy this requirement?

Accepted Answer

Use a hopping window logic where the window size is 15 minutes and the hop (advance) is 5 minutes, producing overlapping periods.

Answer

Use the summarize operator with bin() on 15-minute buckets to create fixed, non-overlapping tumbling windows.

Answer

Use the summarize operator with bin() on 5-minute buckets so every window covers exactly one interval.

Answer

Use the row_window_session() function to group transactions by inactivity gaps between purchases.

Question 56

An IoT platform streams device telemetry into a KQL database table named DeviceEvents, where each device sends multiple status updates per minute. Analysts need a query that returns only the most recent event per device (based on the EventTime column), including all columns from that latest record. Which KQL construct should you use to produce this result efficiently?

Accepted Answer

summarize arg_max(EventTime, *) by DeviceId

Answer

summarize max(EventTime) by DeviceId

Answer

distinct DeviceId, EventTime, Status

Answer

summarize count() by DeviceId, EventTime

Question 57

You store IoT telemetry in an Eventhouse (KQL database) table called DeviceReadings, which contains columns Timestamp (datetime), DeviceId (string), and Temperature (real). A real-time dashboard must display the average temperature per device for each fixed, non-overlapping one-hour interval. Which KQL query correctly produces these hourly aggregates?

Accepted Answer

DeviceReadings | summarize avg(Temperature) by DeviceId, bin(Timestamp, 1h)

Answer

DeviceReadings | summarize avg(Temperature) by DeviceId, Timestamp

Answer

DeviceReadings | summarize avg(Temperature) by DeviceId | bin(Timestamp, 1h)

Answer

DeviceReadings | make-series avg(Temperature) default=0 on Timestamp step 30m by DeviceId

Question 58

A real-time analytics team ingests raw semi-structured device telemetry into a bronze table in an Eventhouse (KQL database). They need each newly ingested row to be automatically parsed, enriched with a derived severity column, and written to a curated silver table with zero additional orchestration and the lowest possible latency after ingestion. The transformation logic is a KQL query. What should they implement?

Accepted Answer

A KQL update policy on the silver table that runs the transformation query against the bronze source table during ingestion

Answer

A scheduled Dataflow Gen2 that reads the bronze table every 5 minutes and appends transformed rows to the silver table

Answer

A materialized view over the bronze table that deduplicates and projects the enriched columns

Answer

A Spark structured streaming job that reads the bronze KQL table and writes the transformed output to the silver table

Question 59

A retail company streams clickstream events into Microsoft Fabric. Analysts need to run interactive ad-hoc queries that filter and aggregate the last 90 days of events with sub-second response times, using time-series operators. The same raw data must also be retained for 7 years at the lowest possible storage cost for occasional compliance audits. Which storage design best meets both requirements?

Accepted Answer

Ingest into an Eventhouse for the hot 90-day window, and use a OneLake availability / export to Delta so older data lands cheaply in a lakehouse for long-term retention.

Answer

Store all 7 years of data in an Eventhouse (KQL database) and query it directly for both interactive analytics and compliance audits.

Answer

Store all data only in a lakehouse Delta table and query it with the SQL analytics endpoint for both interactive and compliance needs.

Answer

Store the 90-day data in a Warehouse using T-SQL and archive older data with a Dataflow Gen2 to a separate Warehouse.

Question 60

A data engineering team is ingesting 40 GB of daily gzip-compressed CSV files from an on-premises SFTP server into a Fabric Lakehouse. The files land in a landing folder, and the team must convert them into partitioned Delta tables by transaction date, apply schema validation, and reject rows with malformed numeric fields. The transformation logic is complex, involving conditional derivations across 30+ columns, and the team includes experienced PySpark developers. Which approach best meets these requirements while keeping the pipeline maintainable and performant at scale?

Accepted Answer

Use a Fabric notebook with PySpark to read the CSV files, apply validation and derivation logic, and write partitioned Delta tables using partitionBy on the transaction date column

Answer

Load the raw CSV files into a KQL database and use update policies to validate and derive columns during ingestion

Answer

Use a Dataflow Gen2 with the Power Query editor to apply all 30+ column derivations and write the result to the Lakehouse

Answer

Use a pipeline Copy activity with a mapping to convert CSV to Delta and rely on the copy schema drift setting to reject malformed rows

Question 61

A data engineering team ingests semi-structured JSON application logs (roughly 2 TB per day) into a Fabric Eventhouse. Analysts need to run ad-hoc exploratory queries that parse nested JSON fields, apply time-based pattern searches over raw text, and compute rolling aggregations across billions of rows with sub-second interactive response times. Which transformation and query engine should the team standardize on for this workload?

Accepted Answer

Kusto Query Language (KQL) against the Eventhouse KQL database

Answer

T-SQL against a Warehouse populated by a nightly pipeline copy

Answer

Dataflow Gen2 with Power Query M transformations

Answer

PySpark notebooks reading the raw JSON from a Lakehouse

Question 62

A retail analytics team lands millions of raw application log lines per hour into an Eventhouse in Microsoft Fabric. Analysts must interactively explore the logs, parse dynamic JSON payloads embedded within each line, and run ad hoc time-series aggregations with sub-second response times over the last 30 days. Which query language and engine should they use for this exploration?

Accepted Answer

Kusto Query Language (KQL) against the Eventhouse KQL database

Answer

T-SQL against a Warehouse populated by copying the logs each hour

Answer

PySpark in a notebook reading the logs from a Lakehouse

Answer

Dataflow Gen2 with Power Query M transformations

Question 63

A data engineer creates a OneLake shortcut in a Fabric lakehouse that points to an external Amazon S3 bucket. Reports built on this data intermittently show outdated values even though the source files in S3 were updated hours ago. The engineer confirms the source data is correct and the shortcut path is valid. What is the most likely cause of the stale results?

Accepted Answer

OneLake shortcut caching is enabled, so previously retrieved files are served from cache until the cache retention period expires.

Answer

The shortcut authentication token has expired, preventing new data from being read.

Answer

The lakehouse SQL analytics endpoint requires a manual metadata sync before new shortcut files are visible in tables.

Answer

S3 cross-region replication is delaying propagation of the updated files to the shortcut.

Question 64

A data engineer creates a OneLake shortcut in the Sales lakehouse that points to a Files folder in a Finance lakehouse located in a different workspace. When a notebook run by a colleague on the Sales team attempts to read data through the shortcut, it fails with a 403 (Unauthorized) error, even though the shortcut was created successfully. The colleague has the Viewer role on the Sales workspace. What is the most likely cause of the error?

Accepted Answer

The colleague lacks permission on the target Finance item; OneLake shortcuts use the calling user's identity and require access to the underlying data source.

Answer

OneLake shortcuts cannot span workspace boundaries, so the shortcut definition is invalid and must be recreated within the same workspace.

Answer

The shortcut cache has expired, and running the OPTIMIZE command on the target table will refresh the credentials.

Answer

The Viewer role on the Sales workspace automatically grants read access to any shortcut target, so the failure indicates a corrupted shortcut path.

Exam domain	Exam weight	Practice
Implement and manage an analytics solution	34%	Practice this topic
Ingest and transform data	33%	Practice this topic
Monitor and optimize an analytics solution	33%	Practice this topic

Microsoft Fabric Data Engineer Associate (DP-700) practice exam & study guide

Start studying DP-700

All DP-700 study resources

DP-700 exam domains

Sample DP-700 questions

Key DP-700 terms

DP-700 frequently asked questions

Implement and manage an analytics solution (34%)

Ingest and transform data (33%)

Monitor and optimize an analytics solution (33%)