Distributed Databases

Q1. A distributed database is:

1. A.A database stored only in the cloud without any local presence
2. B.A database stored on a single machine in one physical location
3. C.A database spread across multiple sites connected by a network✓ Correct
4. D.A backup copy of a database kept for disaster recovery purposes

Report an error in this question

Q2. Data fragmentation in distributed databases means:

1. A.Duplicating the entire database at every site simultaneously
2. B.Compressing data to reduce storage requirements at each site
3. C.Losing data during transfer between sites on the network link
4. D.Dividing a relation into smaller parts stored at different sites✓ Correct

Report an error in this question

Q3. Horizontal fragmentation divides a relation by:

1. A.Rows and columns both
2. B.Rows (tuples) only✓ Correct
3. C.Neither rows nor cols
4. D.Columns (attributes)

Report an error in this question

Q4. Vertical fragmentation divides a relation by:

1. A.Rows (tuples) only
2. B.Columns (attributes)✓ Correct
3. C.Neither of the above
4. D.Columns and rows both

Report an error in this question

Q5. Data replication in distributed databases means:

1. A.Compressing data to reduce transfer size
2. B.Deleting data from all distributed sites
3. C.Encrypting data before sending it across
4. D.Storing copies of data at multiple sites✓ Correct

Report an error in this question

Q6. The transparency goal in distributed databases means:

1. A.Data cannot be distributed across multiple sites
2. B.Users must know all data locations before querying
3. C.Only administrators can query distributed data
4. D.Users should not be aware that data is distributed✓ Correct

Report an error in this question

Q7. Location transparency means:

1. A.Data is always stored locally on the user's own machine
2. B.Users must specify the exact data location in every query
3. C.Users do not need to know where data is physically stored✓ Correct
4. D.No remote access is possible from any site in the system

Report an error in this question

Q8. A distributed DBMS (DDBMS) manages:

1. A.Only local databases without any distribution
2. B.A database distributed across multiple sites✓ Correct
3. C.Only a centralized database on one machine
4. D.Only cloud databases hosted by third parties

Report an error in this question

Q9. Which of the following is an advantage of distributed databases?

1. A.Slower queries being preferred by users
2. B.Higher costs being beneficial overall
3. C.Improved reliability and availability✓ Correct
4. D.Increased complexity as an advantage

Report an error in this question

Q10. Fragmentation transparency means:

1. A.Users are unaware that data is fragmented✓ Correct
2. B.Data cannot be fragmented at any site
3. C.Only fragments can be queried directly
4. D.Users must reassemble fragments manually

Report an error in this question

Q11. The two-phase commit (2PC) protocol ensures:

1. A.Only read operations complete
2. B.No atomicity guarantees at all
3. C.Only local transaction atomicity
4. D.Atomicity of distributed transactions✓ Correct

Report an error in this question

Q12. In the 2PC protocol, the coordinator first sends:

1. A.A prepare (vote) message to all participants✓ Correct
2. B.A commit message directly to all the sites
3. C.An abort message to cancel the transaction
4. D.A query to all participants for data first

Report an error in this question

Q13. If any participant votes No in 2PC, the coordinator:

1. A.Ignores the vote and proceeds with commit
2. B.Sends an abort message to all participants✓ Correct
3. C.Retries the transaction automatically again
4. D.Sends a commit message to all participants

Report an error in this question

Q14. A distributed query involves:

1. A.No data access from any site at all
2. B.Only local data access on one site
3. C.Accessing data from a single table
4. D.Accessing data from multiple sites✓ Correct

Report an error in this question

Q15. The CAP theorem states that a distributed system can guarantee at most:

1. A.Two of three: Consistency, Availability, Partition Tolerance✓ Correct
2. B.None of these properties can be guaranteed in any system
3. C.Only one property out of the three possible in the theorem
4. D.All three properties at the same time without any trade-offs

Report an error in this question

Q16. Semi-join optimization in distributed queries:

1. A.Increases data transfer by sending all records across the link
2. B.Uses full Cartesian products between all fragments in the join
3. C.Reduces data transfer by sending only relevant join attributes✓ Correct
4. D.Eliminates all joins from the distributed query execution plan

Report an error in this question

Q17. Mixed fragmentation combines:

1. A.No fragmentation is used at all
2. B.Only vertical fragmentation types
3. C.Horizontal and vertical fragmentation✓ Correct
4. D.Only horizontal fragmentation types

Report an error in this question

Q18. Full replication means:

1. A.Every site has a complete copy of the entire database✓ Correct
2. B.No replication exists at any site in the system at all
3. C.Only one site has the data and others have no copies
4. D.Data is partially replicated at only a few of sites

Report an error in this question

Q19. The blocking problem in 2PC occurs when:

1. A.No failures occur during the execution of the two-phase commit protocol
2. B.The coordinator fails after sending prepare but before sending the decision✓ Correct
3. C.All participants commit successfully without any failures in the protocol
4. D.The network is fast and reliable with no message delays or lost packets

Report an error in this question

Q20. Distributed deadlock detection is more complex because:

1. A.All transactions are local and never span multiple sites
2. B.Deadlocks can involve transactions at multiple sites✓ Correct
3. C.Deadlocks never occur in distributed database systems
4. D.Only one site has transactions running at any time

Report an error in this question

Q21. The three-phase commit (3PC) protocol improves on 2PC by:

1. A.Using only one phase for the commit decision
2. B.Removing the prepare phase from the protocol
3. C.Adding a pre-commit phase to avoid blocking✓ Correct
4. D.Ignoring all failures that occur at any site

Report an error in this question

Q22. In distributed databases, the global query optimization must consider:

1. A.Only CPU time without considering disk or network access costs
2. B.Only disk I/O costs without considering network or CPU overhead
3. C.Only local processing costs without considering network overhead
4. D.Network communication costs in addition to local processing costs✓ Correct

Report an error in this question

Q23. Paxos is a protocol used for:

1. A.Achieving consensus in distributed systems✓ Correct
2. B.Indexing data for faster retrieval queries
3. C.Query optimization in database management
4. D.Schema design for database normalization

Report an error in this question

Q24. The RAFT consensus algorithm is designed to be:

1. A.Faster than all other protocols in every distributed environment
2. B.A replacement for SQL in distributed relational database systems
3. C.More understandable than Paxos while providing the same guarantees✓ Correct
4. D.A type of index for organizing data in distributed hash tables

Report an error in this question

Q25. In a federated database system:

1. A.No integration is possible between any databases at all
2. B.Autonomous databases cooperate to provide integrated access✓ Correct
3. C.Only one database exists in the entire distributed system
4. D.All databases are identical copies of the same data schema

Report an error in this question

Q26. The eventual consistency model guarantees that:

1. A.No consistency is provided between any of the replicas in the system
2. B.Immediate consistency at all times for every read at every site node
3. C.All reads always return the latest written value at every site instantly
4. D.All replicas will converge to the same value if no new updates are made✓ Correct

Report an error in this question

Q27. The quorum-based protocol requires:

1. A.Qr = Qw = 1 for consistency meaning only one replica needs to respond
2. B.Read quorum (Qr) + Write quorum (Qw) > total replicas (N) for consistency✓ Correct
3. C.No quorum is needed and any single replica can respond to operations
4. D.Qr + Qw less than N for consistency between read and write quorums

Report an error in this question

Q28. Vector clocks in distributed systems are used to:

1. A.Capture causal relationships between events at different sites✓ Correct
2. B.Measure CPU speed and processing capacity at each network node
3. C.Optimize queries by selecting the best execution plan for joins
4. D.Tell wall-clock time accurately across all distributed node sites

Report an error in this question

Q29. Distributed hash tables (DHTs) are used in:

1. A.Peer-to-peer distributed storage systems✓ Correct
2. B.Centralized databases on a single machine
3. C.Single-machine databases without networks
4. D.Only for indexing within one database node

Report an error in this question

Q30. The coordinator selection problem in distributed databases can be solved by:

1. A.Manual intervention always requiring a human administrator
2. B.No algorithm exists for solving the coordinator selection
3. C.Random selection only without any formal algorithmic approach
4. D.Election algorithms like the Bully algorithm or Ring algorithm✓ Correct

Report an error in this question