Questions OnThree Storage Modes
Questions OnThree Storage Modes
• Import Mode
• DirectQuery Mode
• Dual Mode (Composite Mode)
Import mode
• Data is imported and stored in Power BI’s in-
memory engine
• High performance
• Works well with large datasets
• Best suited for complex calculations and
aggregations.
• Consumes more RAM and storage
• Refreshing data requires reloading the entire
dataset,
DirectQuery Mode
• Data remains in the original source, and Power
BI sends queries directly to the database
• No data is stored in Power BI,
• up-to-date data
• Performance depends on the database
• Limited support for complex transformations
and DAX calculations.
Dual Mode (Composite Mode)
• either Import or DirectQuery mode
• Balances performance and real-time data
access
• Frequently used data is cached, while rarely
used data is retrieved on demand.
• Requires careful modeling
• Complexity increases as both modes have
different behaviors.
Modifications to Storage Mode
• Power BI allows switching between storage
modes in Model View:
– Changing Import to DirectQuery: Not directly
possible; requires reconfiguring the dataset.
– Changing DirectQuery to Import: Allowed, but
data will be loaded into Power BI’s memory.
– Using Composite Mode: You can mix Import and
DirectQuery for flexibility.
Important points to note down
• Context is Key: The "best" answer often depends on the specific
scenario outlined in the question. Pay close attention to the details
of the requirements.
• Performance vs. Freshness: The choice between Import and
DirectQuery often involves a trade-off between performance (Import
usually faster) and data freshness (DirectQuery always up-to-date).
• Data Volume: Large data volumes often make DirectQuery a more
practical choice, but aggregation and Dual mode can mitigate
performance issues.
• Data Transformation: Import mode allows for extensive data
transformation within Power Query.
• DAX: DirectQuery may have some limitations with certain DAX
functions, although this is becoming less of an issue.
• Question 1: In which storage mode are tables
solely stored in-memory, and queries are
fulfilled by cached data?
• A. Import
• B. DirectQuery
• C. Dual
• D. Native
• Import
• Question 2: Which storage mode leaves the
data at the data source and retrieves it upon
each query?
• A. Import
• B. DirectQuery
• C. Dual
• D. Live connection
• DirectQuery
• Question 3: Which storage mode allows a
table to act as both Import and DirectQuery,
depending on the query context?
• A. Import
• B. DirectQuery
• C. Dual
• D. Composite
• . Dual
• Question 4: You have a Sales table that needs
near real-time data and a SalesAggregate table
that is updated weekly. Which storage modes
should you use for these tables?
• A. Sales: Import; SalesAggregate: DirectQuery
• B. Sales: DirectQuery; SalesAggregate: Import
• C. Sales: Dual; SalesAggregate: Import
• D. Sales: Import; SalesAggregate: Dual
• B. Sales: DirectQuery; SalesAggregate: Import
• Question 5: Which storage modes will copy
data to Microsoft Power BI? Select all that
apply.
• A. DirectQuery mode
• B. Dual mode
• C. Import mode
• B. Dual mode; C. Import mode
• You plan to add data to Power BI Desktop from a
new data source. You are evaluating whether you
should use the DirectQuery storage mode or the
Import storage mode.
• What are two benefits of using DirectQuery
instead of Import? Each correct answer presents
a complete solution.
• What is the maximum number of data
refreshes per day?
• Select only one answer.
• 1
• 4
• 8
• 24
• 8
• What is the number of consecutive refresh
failures of DS1 that will automatically disable
the refresh schedule?
• 2
• 4
• 16
• 64
• 4
• What is the maximum number of refreshes
that can be scheduled each day?
• 1
• 8
• 16
• 48
• 48
• A business intelligence (BI) developer creates a dataflow in Power BI that uses
DirectQuery to access tables from an on-premises Microsoft SQL server. The
Enhanced Dataflows Compute Engine is turned on for the dataflow.
You need to use the dataflow in a report. The solution must meet the
following requirements:
✑ Minimize online processing operations.
✑ Minimize calculation times and render times for visuals.
✑ Include data from the current year, up to and including the previous day.
What should you do?
Model1 must show updated data within two minutes of the data
being updated in DB1.