A data warehouse has a large FactSales table and a smaller DimProduct table. A reporting query joins these tables and aggregates sales by product category. The query is slow. You need to improve performance without changing the reporting query. What should you do?

Create an index on DimProduct.ProductID.

Use a Dataflow Gen2 to pre-join the tables and load into a new table.

Create an aggregated table that pre-joins FactSales and DimProduct and stores sales by category.

Use a KQL query to perform the join and aggregation.

What is the primary benefit of using PySpark for batch data transformation in Microsoft Fabric?

It automatically creates Power BI reports

It provides a low-code visual interface

It enables distributed processing of large datasets across a cluster

It is the only language supported in Fabric notebooks

A nightly batch process loads customer records into a Lakehouse table. Some records are duplicated due to a source system issue. The duplicates are identified by the CustomerID column. The data engineer needs to ensure that only the latest record for each CustomerID is kept, based on a LastUpdated timestamp. The transformation should be done in a notebook using PySpark. What should the engineer do?

Use distinct() on the DataFrame.

Use a window function with row_number() partitioned by CustomerID ordered by LastUpdated desc, then filter for row number 1.

Use dropDuplicates with a subset of CustomerID and keep the first occurrence.

Use groupBy CustomerID and agg max(LastUpdated).

In a batch pipeline, what is the purpose of a watermark in the context of late-arriving data?

To define a boundary for how late data can arrive and still be included

What is a common approach to handle late-arriving data in a batch processing workflow?

Discarding all late-arriving data

Using watermarking and reprocessing windows

Processing late data in real-time only

Ignoring late data until the next full load

Microsoft Certified:Fabric Data Engineer Associate

Domain 2Objective 2

Ingest and Transform Batch Data DP-700 Practice Questions (Page 2)

Part of the Ingest and transform data domain, which accounts for 30–35% of the DP-700 exam. Microsoft does not publish an official question count, but from its 100-minute exam (~40–65 total, ~12–23 in this domain), expect 4–8 from this objective — we provide 48 practice questions to prepare you well beyond it. (estimate)

48questions here

10free pages

13concepts

30–35%of the exam

Questions 6–10

6expert · hard
A data warehouse has a large FactSales table and a smaller DimProduct table. A reporting query joins these tables and aggregates sales by product category. The query is slow. You need to improve performance without changing the reporting query. What should you do?
Select an answer first
7foundation · easy
What is the primary benefit of using PySpark for batch data transformation in Microsoft Fabric?
Select an answer first
8application · medium
A nightly batch process loads customer records into a Lakehouse table. Some records are duplicated due to a source system issue. The duplicates are identified by the CustomerID column. The data engineer needs to ensure that only the latest record for each CustomerID is kept, based on a LastUpdated timestamp. The transformation should be done in a notebook using PySpark. What should the engineer do?
Select an answer first
9foundation · easy
In a batch pipeline, what is the purpose of a watermark in the context of late-arriving data?
Select an answer first
10foundation · easy
What is a common approach to handle late-arriving data in a batch processing workflow?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to DP-700

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Microsoft. “DP-700” is a trademark of its owner, used for identification only.