MCQ Collection
Data Warehousing and Business Intelligence MCQs
Practice Data Warehousing and Business Intelligence questions with answers and explanations.
Choose an option to check your answer.
A.
Splitting a measure into numerator and denominator
B.
Dividing one logical table into separately managed physical segments based on a key
C.
Creating an unrelated copy of every dimension
D.
Moving charts to different dashboard pages
Show Answer
Correct Answer: B. Dividing one logical table into separately managed physical segments based on a key
Explanation:
Partitioning keeps one table interface while organizing data into ranges or groups.
Warehouse fact tables are commonly partitioned by date.
Choose an option to check your answer.
A.
A list of dashboard users
B.
A record of source inserts, updates, and deletes that occurred after a point in time
C.
A prebuilt star schema
D.
A method for choosing chart colors
Show Answer
Correct Answer: B. A record of source inserts, updates, and deletes that occurred after a point in time
Explanation:
CDC exposes data changes so the warehouse can load them incrementally.
It is more precise than repeatedly scanning the entire source table.
Choose an option to check your answer.
A.
Failures are hidden from operators
B.
A failed workflow can resume from a safe checkpoint instead of repeating all work
C.
Source data is automatically corrected
D.
Every package runs in parallel
Show Answer
Correct Answer: B. A failed workflow can resume from a safe checkpoint instead of repeating all work
Explanation:
Checkpoints and batch states reduce recovery time after failures.
Restart logic must avoid partial duplicates and maintain consistency.
Choose an option to check your answer.
A.
A dimension row inserted by source users
B.
A value defined by an expression rather than stored directly as a base fact
C.
A filegroup containing cube backups
D.
A nonclustered index on a fact table
Show Answer
Correct Answer: B. A value defined by an expression rather than stored directly as a base fact
Explanation:
Calculated members derive reusable analytical results such as margin or year-to-date sales.
The model evaluates the expression from underlying stored measures.
Choose an option to check your answer.
A.
All detail must fit in client memory
B.
Scalability and relational freshness depend on database query performance
C.
Cube processing always duplicates every source row
D.
Users cannot drill through to detail
Show Answer
Correct Answer: B. Scalability and relational freshness depend on database query performance
Explanation:
ROLAP queries relational tables rather than storing all detail multidimensionally.
This supports large or frequently changing data but requires strong relational tuning.
Choose an option to check your answer.
A.
A Customer dimension containing names and addresses
B.
A staging table containing rejected rows
C.
A table recording student attendance using only dimension keys and no numeric measure
D.
A summary table containing monthly revenue
Show Answer
Correct Answer: C. A table recording student attendance using only dimension keys and no numeric measure
Explanation:
Factless facts record the occurrence of an event or coverage relationship.
The event count itself can be derived by counting rows.
Choose an option to check your answer.
A.
A Type 1 overwrite of both representatives
B.
Removing the sales amount from the fact table
C.
Allocation weights whose total equals the full fact amount
D.
Creating unrelated copies of the fact
Show Answer
Correct Answer: C. Allocation weights whose total equals the full fact amount
Explanation:
Bridge rows can carry weights that distribute one fact across multiple members.
Correct weighting preserves the original total during aggregation.
Choose an option to check your answer.
A.
All partitions are merged before each query
B.
The database removes dimension keys
C.
The optimizer reads only partitions relevant to the query predicate
D.
The ETL process deletes old reports
Show Answer
Correct Answer: C. The optimizer reads only partitions relevant to the query predicate
Explanation:
A predicate aligned with the partition key lets irrelevant partitions be skipped.
This reduces I/O for date-bounded analytical queries.
Choose an option to check your answer.
A.
The largest fact amount is removed from each batch
B.
Every source column is converted to a percentage
C.
The last successfully processed timestamp or sequence is saved for the next extraction
D.
The dashboard displays only peak values
Show Answer
Correct Answer: C. The last successfully processed timestamp or sequence is saved for the next extraction
Explanation:
A watermark defines the boundary between previously loaded and new records.
It must be updated only after a successful, controlled load.
Choose an option to check your answer.
A.
All tables can always load in any order
B.
Dependencies improve dashboard decoration
C.
Some loads require prerequisite data or successful upstream completion
D.
They prevent source systems from recording transactions
Show Answer
Correct Answer: C. Some loads require prerequisite data or successful upstream completion
Explanation:
Dimensions, facts, aggregates, and semantic models often depend on earlier tasks.
Orchestration ensures correct sequence and prevents incomplete downstream results.
Choose an option to check your answer.
A.
A transaction log and backup file
B.
A source-system primary key only
C.
A target, status, and trend definition
D.
A list of all cube partitions
Show Answer
Correct Answer: C. A target, status, and trend definition
Explanation:
A KPI compares performance with a business goal and indicates direction or condition.
Status and trend help users interpret the value rather than viewing it alone.
Choose an option to check your answer.
A.
To avoid using dimensions entirely
B.
To process transactions directly in a cube
C.
To combine fast multidimensional summaries with relational storage for detailed data
D.
To store only dashboard screenshots
Show Answer
Correct Answer: C. To combine fast multidimensional summaries with relational storage for detailed data
Explanation:
HOLAP balances MOLAP-style aggregate speed with ROLAP-style detailed scalability.
The architecture is useful when detail volume is too large for full MOLAP storage.