1.

Transaction Properties (ACID)

Q1. A transaction modifies two accounts: it debits ₹500 from Account A and credits ₹500 to
Account B. Due to a system crash after debiting Account A but before crediting Account B, the
database is left inconsistent. Which ACID property is violated?

●​ A) Atomicity​

●​ B) Consistency​

●​ C) Isolation​

●​ D) Durability​

Answer: A) Atomicity

Explanation: Atomicity ensures that all operations within a transaction are completed; if not, the
transaction is aborted.

2. Concurrency Control and Schedules

Q2. Given the following schedule:Courses at Washington University+1Sanfoundry+1

●​ T1: Read(A), Write(A)​

●​ T2: Read(A), Write(A)Courses at Washington University+3Testbook+3Testbook+3​

Assuming no concurrency control, what anomaly can occur?

●​ A) Lost Update​

●​ B) Dirty Read​

●​ C) Non-repeatable Read​

●​ D) Phantom Read​

Answer: A) Lost Update

Explanation: Both transactions read and write the same data item without isolation, leading to
one update being overwritten.

3. Serializability

Q3. Which of the following schedules is conflict-serializable?

●​ A) T1: Read(A), Write(A); T2: Read(B), Write(B)​

●​ B) T1: Read(A), Write(A); T2: Read(A), Write(A)​

●​ C) T1: Read(A), Write(B); T2: Read(B), Write(A)​

●​ D) T1: Read(A), Write(B); T2: Read(B), Write(C)​

Answer: A) T1: Read(A), Write(A); T2: Read(B), Write(B)

Explanation: Since T1 and T2 operate on different data items, their operations do not conflict,
making the schedule conflict-serializable.

4. Two-Phase Locking (2PL)

Q4. In the Two-Phase Locking protocol, which phase allows the release of locks?

●​ A) Growing Phase​

●​ B) Shrinking Phase​

●​ C) Both Phases​

●​ D) Neither Phase​

Answer: B) Shrinking Phase

Explanation: In 2PL, a transaction releases locks only during the shrinking phase after
acquiring all needed locks in the growing phase.5. Deadlock Detection

Q5. Which of the following is a necessary condition for a deadlock to occur?

 ●​ A) Mutual Exclusion​

●​ B) Preemption​

●​ C) Circular Wait​

●​ D) All of the above​

Answer: D) All of the above

Explanation: Deadlock requires mutual exclusion, hold and wait, no preemption, and circular
wait conditions.

6. Timestamp-Based Concurrency Control

Q6. In timestamp ordering, if a transaction T1 has a timestamp TS(T1) and attempts to write to
data item Q, but TS(T1) < read_TS(Q), what action is taken?

●​ A) The write is allowed​

●​ B) The transaction is aborted​

●​ C) The transaction is delayed​

●​ D) The transaction is committed​

Answer: B) The transaction is aborted

Explanation: If TS(T1) < read_TS(Q), it indicates that a younger transaction has already read
the data, so T1 must be aborted to maintain consistency.

7. Recovery Techniques

Q7. In deferred update recovery technique, when are the changes written to the database?

●​ A) Immediately after each operation​

●​ B) After the transaction starts​

 ●​ C) After the transaction commits​

●​ D) At scheduled intervals​

Answer: C) After the transaction commits

Explanation: In deferred update, all changes are recorded in the log and applied to the
database only after the transaction commits.

8. Immediate Update Recovery

Q8. Which of the following is true about immediate update recovery technique?

●​ A) Changes are applied to the database after commit​

●​ B) Undo and redo operations may be required​

●​ C) Only redo operations are needed​

●​ D) No logging is necessary​

Answer: B) Undo and redo operations may be required

Explanation: Since changes are applied before commit, both undo (for uncommitted
transactions) and redo (for committed transactions) may be necessary.

9. Shadow Paging

Q9. What is the primary advantage of shadow paging in recovery?

●​ A) Requires complex logging​

●​ B) Eliminates the need for undo operations​

●​ C) Increases transaction throughput​

●​ D) Simplifies concurrency control​

Answer: B) Eliminates the need for undo operations

Explanation: Shadow paging maintains two page tables, allowing the system to revert to the
shadow copy in case of failure, thus avoiding undo operations.

10. Schedule Recoverability

Q10. Which of the following schedules is recoverable?

●​ A) T1 commits before T2 reads data written by T1​

●​ B) T2 reads data written by T1, then T1 commits​

●​ C) T2 commits before T1, which it read from, commits​

●​ D) T2 reads data written by T1, then T1 aborts​

Answer: B) T2 reads data written by T1, then T1 commits

Explanation: A schedule is recoverable if transactions commit only after the transactions whose
changes they read have committed.

11.

If T1 reads X and then T2 writes X before T1 completes, what kind of anomaly can occur?

●​ A) Lost Update​

●​ B) Dirty Read​

●​ C) Inconsistent Retrieval​

●​ D) Write Skew​

Answer: B) Dirty Read

12.

A schedule where T2 reads a value written by T1, but T1 later aborts, is considered:
 ●​ A) Recoverable​

●​ B) Non-recoverable​

●​ C) Serializable​

●​ D) Strict​

Answer: B) Non-recoverable

13.

Which locking protocol ensures serializability but does not guarantee deadlock freedom?

●​ A) Basic Two-Phase Locking​

●​ B) Wait-die​

●​ C) Timeout-based​

●​ D) Timestamp ordering​

Answer: A) Basic Two-Phase Locking

14.

Transaction T1 acquires a shared lock on data item A and T2 requests an exclusive lock on A.
What will happen?

●​ A) T2 gets the lock immediately​

●​ B) T2 is blocked until T1 releases the lock​

●​ C) T1 is aborted​

●​ D) Both execute simultaneously​

Answer: B) T2 is blocked until T1 releases the lock

15.

Which of the following is true for the Wait-Die scheme?

●​ A) Older transaction waits for younger​

●​ B) Older transaction is aborted​

●​ C) Younger transaction waits for older​

●​ D) Deadlock is not avoided​

Answer: A) Older transaction waits for younger

16.

Which type of schedule is always serializable?

●​ A) Cascadeless​

●​ B) Conflict-serializable​

●​ C) Strict​

●​ D) View-serializable​

Answer: C) Strict

17.

If a transaction must release all its locks only after it finishes, it follows:

●​ A) Conservative 2PL​

●​ B) Strict 2PL​

●​ C) Basic 2PL​
 ●​ D) Rigorous 2PL​

Answer: D) Rigorous 2PL

18.

If a transaction writes to a data item and another transaction reads it before the first one
commits, it’s called:

●​ A) Dirty Read​

●​ B) Non-repeatable Read​

●​ C) Phantom Read​

●​ D) Write Skew​

Answer: A) Dirty Read

19.

Which concurrency control method assigns a unique timestamp to each transaction to maintain
serializability?

●​ A) Wait-die​

●​ B) Timestamp Ordering​

●​ C) Graph-based protocol​

●​ D) Deadlock Prevention​

Answer: B) Timestamp Ordering

20.

Which condition is not required for a deadlock?

 ●​ A) Circular Wait​

●​ B) Hold and Wait​

●​ C) Resource Preemption​

●​ D) Mutual Exclusion​

Answer: C) Resource Preemption

21.

If two transactions are holding locks and waiting for each other's data items, the system is said
to be in:

●​ A) Starvation​

●​ B) Deadlock​

●​ C) Race Condition​

●​ D) Serial Execution​

Answer: B) Deadlock

22.

Which recovery method uses before and after images in the log?

●​ A) Shadow Paging​

●​ B) Deferred Update​

●​ C) Immediate Update​

●​ D) Write-Ahead Logging​

Answer: C) Immediate Update

23.

In which technique is a new page table created during the transaction, and the original remains
unchanged?

●​ A) Deferred Update​

●​ B) Shadow Paging​

●​ C) Undo Logging​

●​ D) Redo Logging​

Answer: B) Shadow Paging

24.

Which concurrency anomaly occurs when a transaction reads the same row twice and gets
different values?

●​ A) Dirty Read​

●​ B) Non-repeatable Read​

●​ C) Phantom Read​

●​ D) Lost Update​

Answer: B) Non-repeatable Read

25.

Which schedule avoids cascading aborts?

●​ A) Non-recoverable schedule​

●​ B) Recoverable schedule​

●​ C) Cascadeless schedule​
 ●​ D) Serializable schedule​

Answer: C) Cascadeless schedule

26.

In timestamp ordering, what happens if a transaction tries to write to a data item that has
already been read by a newer transaction?

●​ A) The write proceeds​

●​ B) The transaction waits​

●​ C) The transaction is rolled back​

●​ D) Timestamp is updated​

Answer: C) The transaction is rolled back

27.

If a transaction fails after some of its changes have been made to the database, what must the
recovery manager do?

●​ A) Perform redo only​

●​ B) Perform undo only​

●​ C) Perform redo and undo​

●​ D) Ignore the failure​

Answer: C) Perform redo and undo

28.

Which of the following is a disadvantage of Shadow Paging?

 ●​ A) No atomicity​

●​ B) Poor concurrency​

●​ C) High overhead for page copying​

●​ D) Requires undo/redo​

Answer: C) High overhead for page copying

29.

Which lock is required before reading a data item?

●​ A) Exclusive lock​

●​ B) Shared lock​

●​ C) Write lock​

●​ D) Phantom lock​

Answer: B) Shared lock

30.

A transaction reads a set of rows that satisfy a WHERE clause. Another transaction inserts a
new row that would have satisfied that clause. This is:

●​ A) Dirty Read​

●​ B) Phantom Read​

●​ C) Lost Update​

●​ D) Write Skew​

Answer: B) Phantom Read

31.

Which of the following schedules is both conflict-serializable and recoverable?

●​ A) T1 reads A, T2 writes A, T1 commits​

●​ B) T2 reads A, T1 writes A, T1 commits, T2 commits​

●​ C) T1 writes A, T2 reads A, T1 commits, T2 commits​

●​ D) T1 reads A, T2 writes A, T2 commits, T1 commits​

Answer: C) T1 writes A, T2 reads A, T1 commits, T2 commits

32.

Which of the following protocols avoids deadlock by not acquiring any locks?

●​ A) Timestamp ordering​

●​ B) Two-phase locking​

●​ C) Wait-die​

●​ D) Wound-wait​

Answer: A) Timestamp ordering

33.

In wound-wait scheme, what happens when a younger transaction requests a lock held by an
older one?

●​ A) The older one waits​

●​ B) The younger is aborted​

 ●​ C) The younger waits​

●​ D) Deadlock occurs​

Answer: B) The younger is aborted

34.

A transaction performing only read operations doesn't require:

●​ A) Shared locks​

●​ B) Exclusive locks​

●​ C) Timestamp​

●​ D) Deadlock avoidance​

Answer: B) Exclusive locks

35.

Which of the following ensures no dirty reads, non-repeatable reads, or phantom reads?

●​ A) Read Uncommitted​

●​ B) Read Committed​

●​ C) Repeatable Read​

●​ D) Serializable​

Answer: D) Serializable

36.

In which case does a redo operation become necessary during recovery?

 ●​ A) Transaction has not committed​

●​ B) Log has only before images​

●​ C) After images are present in log​

●​ D) Deferred update technique is used​

Answer: C) After images are present in log

37.

Which lock mode allows multiple transactions to access the same data item simultaneously?

●​ A) Exclusive​

●​ B) Shared​

●​ C) Intent Exclusive​

●​ D) Recursive​

Answer: B) Shared

38.

The principle of write-ahead logging ensures:

●​ A) Undo is done before redo​

●​ B) Redo is done before undo​

●​ C) Log is written before database is updated​

●​ D) All locks are released before commit​

Answer: C) Log is written before database is updated

39.

In strict 2PL, when are all locks released?

●​ A) Before commit​

●​ B) During execution​

●​ C) After commit​

●​ D) After acquiring all locks​

Answer: C) After commit

40.

Which technique guarantees atomicity but has poor performance for long transactions?

●​ A) Deferred Update​

●​ B) Shadow Paging​

●​ C) Write-Ahead Logging​

●​ D) 2PL​

Answer: B) Shadow Paging

41.

Transaction T1 writes to X. Before T1 commits, T2 reads X. What property is violated?

●​ A) Atomicity​

●​ B) Isolation​

●​ C) Durability​

●​ D) Consistency​
Answer: B) Isolation​
Explanation: T2 reading uncommitted data from T1 violates isolation, leading to a dirty read
anomaly.

42.

Which recovery mechanism avoids undo operations altogether?

●​ A) Immediate Update​

●​ B) Shadow Paging​

●​ C) Write-ahead Logging​

●​ D) Checkpointing​

Answer: B) Shadow Paging​

Explanation: Shadow Paging preserves the original database page; if failure occurs, no undo
is needed since changes are only written upon commit.

43.

A schedule where every transaction reads only committed data and commits only after all others
it depends on have committed is:

●​ A) Cascading​

●​ B) Strict​

●​ C) Recoverable​

●​ D) Serializable​

Answer: C) Recoverable​
Explanation: In a recoverable schedule, transactions commit only after the transactions they
read from commit.

44.
Which protocol guarantees both serializability and deadlock freedom?

●​ A) Strict 2PL​

●​ B) Basic 2PL​

●​ C) Timestamp Ordering​

●​ D) Graph-based Locking​

Answer: C) Timestamp Ordering​

Explanation: Timestamp ordering uses logical timestamps to ensure transactions execute in
a serializable and deadlock-free manner.

45.

In the deferred update technique, when are changes applied to the database?

●​ A) As soon as the transaction executes​

●​ B) After log writing​

●​ C) Only after the transaction commits​

●​ D) Before validation​

Answer: C) Only after the transaction commits​

Explanation: In deferred update, changes are delayed until commit, simplifying undo during
failure.

46.

Which of the following prevents phantom reads?

●​ A) Exclusive locks​

●​ B) Strict 2PL​

●​ C) Predicate locking​
 ●​ D) Shared locks​

Answer: C) Predicate locking​

Explanation: Predicate locks lock the result of a query condition, preventing other
transactions from adding/removing matching rows (phantoms).

47.

Which statement is true about Strict 2PL?

●​ A) All locks are released after the transaction ends​

●​ B) Read locks are released earlier than write locks​

●​ C) Locks are never released​

●​ D) Only read locks are used​

Answer: A) All locks are released after the transaction ends​

Explanation: In Strict 2PL, locks are held until the transaction either commits or aborts.

48.

Which of the following does not cause a schedule to be unrecoverable?

●​ A) T2 reads data written by uncommitted T1​

●​ B) T1 aborts after T2 reads its write​

●​ C) T2 commits before T1 commits​

●​ D) T2 reads unmodified data from T1​

Answer: D) T2 reads unmodified data from T1​

Explanation: Reading unmodified data does not create dependency or violate recoverability.

49.
When two transactions conflict on a data item and their operations are not reordered to match a
serial execution, the schedule is:

●​ A) Serializable​

●​ B) Conflict-serializable​

●​ C) Non-serializable​

●​ D) View-serializable​

Answer: C) Non-serializable​
Explanation: If operations cannot be reordered without violating dependency, the schedule is
non-serializable.

50.

Which of the following best describes durability in transactions?

●​ A) Changes are never lost once transaction starts​

●​ B) Changes are visible to other transactions immediately​

●​ C) Changes persist after commit even during a crash​

●​ D) Transaction can be undone anytime​

Answer: C) Changes persist after commit even during a crash​

Explanation: Durability ensures committed changes survive system crashes.

51.

Which component is responsible for transaction recovery?

●​ A) Scheduler​

●​ B) Buffer Manager​

●​ C) Transaction Manager​
 ●​ D) Recovery Manager​

Answer: D) Recovery Manager​

Explanation: The Recovery Manager tracks logs and ensures atomicity/durability after
crashes.

52.

Which schedule violates consistency but still appears conflict-serializable?

●​ A) Dirty read​

●​ B) Lost update​

●​ C) Write skew​

●​ D) Non-repeatable read​

Answer: C) Write skew​

Explanation: Write skew can occur in conflict-serializable schedules, breaking consistency.

53.

In timestamp ordering, a transaction T issues a read(X), but TS(T) < WTS(X). What happens?

●​ A) Read proceeds​

●​ B) T is aborted​

●​ C) X is locked​

●​ D) X is updated​

Answer: B) T is aborted​
Explanation: Since T is older and X has been written by a newer transaction, T must abort to
avoid inconsistency.
54.

Which type of locking strategy can be used to detect rather than prevent deadlock?

●​ A) Timeout-based​

●​ B) Wait-for graph​

●​ C) Wound-wait​

●​ D) Wait-die​

Answer: B) Wait-for graph​

Explanation: Wait-for graph detects cycles to detect deadlocks after they occur.

55.

Which isolation level allows dirty reads?

●​ A) Read Committed​

●​ B) Read Uncommitted​

●​ C) Repeatable Read​

●​ D) Serializable​

Answer: B) Read Uncommitted​

Explanation: Read Uncommitted allows a transaction to read data from uncommitted
transactions.

56.

A system uses undo logging. After a crash, what must be done?

●​ A) Redo all committed transactions​

●​ B) Undo uncommitted transactions​

 ●​ C) Redo uncommitted transactions​

●​ D) Clear the log​

Answer: B) Undo uncommitted transactions​

Explanation: In undo logging, uncommitted updates are rolled back.

57.

Which of the following is not ensured by the two-phase locking protocol?

●​ A) Serializability​

●​ B) Deadlock freedom​

●​ C) Isolation​

●​ D) Concurrency control​

Answer: B) Deadlock freedom​

Explanation: 2PL ensures serializability but does not prevent deadlocks.

58.

Transaction T is older and tries to write to a locked data item held by a younger transaction. In
wound-wait, what will happen?

●​ A) T waits​

●​ B) Younger transaction is aborted​

●​ C) Deadlock​

●​ D) Timeout​

Answer: B) Younger transaction is aborted​

Explanation: In wound-wait, the older transaction "wounds" the younger one to prevent
deadlock.
59.

In immediate update systems, which operation must occur before updating the database?

●​ A) Release of locks​

●​ B) Log record written​

●​ C) Checkpointing​

●​ D) Undo all previous operations​

Answer: B) Log record written​

Explanation: Write-ahead logging requires log records be written before database updates in
immediate update systems.

60.

Which recovery approach simplifies crash handling but has high space overhead?

●​ A) Immediate Update​

●​ B) Deferred Update​

●​ C) Shadow Paging​

●​ D) Undo-Redo Logging​

Answer: C) Shadow Paging​

Explanation: Shadow Paging eliminates undo/redo but requires page copies, increasing
space usage.

61.

Which operation sequence violates conflict serializability?

●​ A) T1: R(X), T2: R(X)​

 ●​ B) T1: W(X), T2: R(X)​

●​ C) T1: R(X), T2: W(X)​

●​ D) T1: W(X), T2: W(X)​

Answer: D) T1: W(X), T2: W(X)​

Explanation: Two write operations on the same item by different transactions create a
conflict and can violate serializability if the order is not preserved.

62.

Which of the following is most likely to cause a write skew?

●​ A) Two-phase locking​

●​ B) Serializable isolation​

●​ C) Snapshot isolation​

●​ D) Read committed isolation​

Answer: C) Snapshot isolation​

Explanation: Snapshot isolation can lead to write skew, a phenomenon where concurrent
transactions see consistent snapshots but their combined effect breaks integrity.

63.

In strict schedules, what is always true?

●​ A) Read and write locks are acquired before use​

●​ B) Unlocking happens after transaction ends​

●​ C) All reads are allowed before writes​

●​ D) Transactions do not use locks​

Answer: B) Unlocking happens after transaction ends​
Explanation: In strict schedules, locks are held until commit/abort, preventing cascading
aborts.

64.

Which of the following is a non-conflicting operation?

●​ A) T1: W(X), T2: W(X)​

●​ B) T1: R(X), T2: W(X)​

●​ C) T1: R(X), T2: R(X)​

●​ D) T1: W(X), T2: R(X)​

Answer: C) T1: R(X), T2: R(X)​

Explanation: Read-read operations on the same data item do not conflict and can execute
concurrently.

65.

Which property ensures that a failure during a transaction doesn't leave the database in an
inconsistent state?

●​ A) Durability​

●​ B) Isolation​

●​ C) Consistency​

●​ D) Atomicity​

Answer: D) Atomicity​
Explanation: Atomicity ensures that either all or none of the transaction executes, avoiding
partial changes.

66.
A transaction reads the same item twice but gets different values. This violates:

●​ A) Durability​

●​ B) Repeatable Read​

●​ C) Serializable​

●​ D) Deadlock​

Answer: B) Repeatable Read​

Explanation: Repeatable read ensures that once a transaction reads an item, it sees the
same value throughout.

67.

Which anomaly occurs when two transactions read the same data and both update based on
the original value?

●​ A) Lost update​

●​ B) Dirty read​

●​ C) Phantom read​

●​ D) Inconsistent read​

Answer: A) Lost update​

Explanation: In a lost update, one transaction’s update gets overwritten by another due to
lack of synchronization.

68.

Which one of the following is a prevention-based deadlock handling strategy?

●​ A) Wait-for graph​

●​ B) Timeout​
 ●​ C) Wait-die​

●​ D) Deadlock detection​

Answer: C) Wait-die​
Explanation: Wait-die is a deadlock prevention technique using timestamps. Younger
transactions abort rather than wait for older ones.

69.

In which case will a transaction be rolled back in timestamp ordering?

●​ A) TS(T) > WTS(X)​

●​ B) TS(T) < RTS(X) for a write​

●​ C) TS(T) > RTS(X) for a read​

●​ D) TS(T) = TS(X)​

Answer: B) TS(T) < RTS(X) for a write​

Explanation: A write by an older transaction after a newer transaction has read the data is
not allowed. Hence, rollback.

70.

Which is a major drawback of the two-phase locking protocol?

●​ A) Allows dirty reads​

●​ B) May lead to starvation​

●​ C) Can cause deadlock​

●​ D) Not serializable​

Answer: C) Can cause deadlock​

Explanation: Though 2PL ensures serializability, it may cause deadlock due to waiting
dependencies.
71.

Which technique ensures both serializability and recoverability?

●​ A) Strict 2PL​

●​ B) Read committed​

●​ C) Shadow Paging​

●​ D) Snapshot Isolation​

Answer: A) Strict 2PL​

Explanation: Strict 2PL enforces both conflict serializability and recoverability by holding
locks until the end.

72.

Which type of schedule permits cascading rollbacks?

●​ A) Recoverable schedule​

●​ B) Strict schedule​

●​ C) Cascadeless schedule​

●​ D) Non-recoverable schedule​

Answer: A) Recoverable schedule​

Explanation: Recoverable schedules allow cascading rollbacks but ensure that no
transaction commits if it reads from uncommitted ones.

73.

In a serial schedule, the transactions are executed:

●​ A) Concurrently​
 ●​ B) Overlapping with some delay​

●​ C) One after the other without interleaving​

●​ D) Randomly​

Answer: C) One after the other without interleaving​

Explanation: A serial schedule executes transactions in a strict sequence — no interleaving.

74.

The write-ahead logging protocol ensures:

●​ A) Logs are written after data pages​

●​ B) Logs are written only on commit​

●​ C) Logs are written before data is updated​

●​ D) No logging is needed if rollback occurs​

Answer: C) Logs are written before data is updated​

Explanation: WAL guarantees that the system can recover using logs in case of failure.

75.

What is the checkpoint in transaction recovery?

●​ A) A command to start a transaction​

●​ B) A point to restart log scanning after crash​

●​ C) A temporary data buffer​

●​ D) Undo buffer flush​

Answer: B) A point to restart log scanning after crash​

Explanation: Checkpointing reduces recovery time by marking a safe point in logs.
76.

Which statement about phantom reads is true?

●​ A) They involve reading stale values​

●​ B) They involve duplicate reads​

●​ C) They occur when new rows satisfy the WHERE clause​

●​ D) They occur during dirty writes​

Answer: C) They occur when new rows satisfy the WHERE clause​
Explanation: Phantom reads occur when a query returns a different set of rows due to inserts
by another transaction.

77.

Which one is NOT a valid serializable schedule?

●​ A) One that results in same final DB state as a serial one​

●​ B) One where all operations can be reordered into a serial order​

●​ C) One with no conflicts between any operations​

●​ D) One where reads always happen before writes​

Answer: D) One where reads always happen before writes​

Explanation: Serializable schedules are not defined by read-before-write ordering but by
conflict-free reordering.

78.

Which one of the following is least likely to lead to a deadlock?

●​ A) Circular waits​
 ●​ B) Waiting due to exclusive locks​

●​ C) Transactions locking resources without releasing​

●​ D) Read-only transactions​

Answer: D) Read-only transactions​

Explanation: Read-only transactions use shared locks and typically don't block others, so
deadlocks are rare.

79.

Which concurrency control mechanism supports greater concurrency but risks

inconsistencies?

●​ A) Serializable​

●​ B) Repeatable Read​

●​ C) Read Committed​

●​ D) Read Uncommitted​

Answer: D) Read Uncommitted​

Explanation: Read Uncommitted allows maximum concurrency but also dirty reads and
anomalies.

80.

What’s the key difference between view serializability and conflict serializability?

●​ A) Conflict serializability is stricter​

●​ B) View serializability requires locking​

●​ C) Conflict serializability allows more schedules​

●​ D) Both are identical​

Answer: A) Conflict serializability is stricter​
Explanation: All conflict-serializable schedules are view-serializable, but not vice versa.

81.

Which schedule below may be non-recoverable?

●​ A) T1: W(X), T2: R(X), T1: Commit, T2: Commit​

●​ B) T1: W(X), T2: R(X), T2: Commit, T1: Abort​

●​ C) T1: R(X), T2: W(X), T1: Commit, T2: Commit​

●​ D) T1: W(X), T2: R(Y), T1: Commit, T2: Commit​

Answer: B) T1: W(X), T2: R(X), T2: Commit, T1: Abort​

Explanation: T2 reads uncommitted data from T1 and commits before T1, violating
recoverability.

82.

Which of the following ensures no cascading rollbacks?

●​ A) Recoverable schedule​

●​ B) Cascadeless schedule​

●​ C) View-serializable schedule​

●​ D) Deadlock-free schedule​

Answer: B) Cascadeless schedule​

Explanation: In cascadeless schedules, transactions read only committed data, avoiding
cascading aborts.

83.

Which operation leads to a dirty read?

 ●​ A) A transaction reading committed data from another​

●​ B) A transaction reading from its own writes​

●​ C) A transaction reading data written by an uncommitted transaction​

●​ D) A transaction writing without checking locks​

Answer: C) A transaction reading data written by an uncommitted transaction​

Explanation: This can lead to incorrect decisions and rollback propagation.

84.

In deferred update recovery, changes are:

●​ A) Written immediately to disk​

●​ B) Kept in memory until commit​

●​ C) Reflected in log after every write​

●​ D) Applied partially on abort​

Answer: B) Kept in memory until commit​

Explanation: Deferred update avoids writing to the DB until commit, making undo
unnecessary.

85.

In which situation is UNDO necessary?

●​ A) All updates are deferred​

●​ B) The transaction commits​

●​ C) A failure occurs before commit​

●​ D) A checkpoint is created​
Answer: C) A failure occurs before commit​
Explanation: UNDO is needed to rollback uncommitted changes after a crash or abort.

86.

What’s the impact of phantom reads?

●​ A) Same row read multiple times​

●​ B) Row value changes during a transaction​

●​ C) New rows appear in re-executed queries​

●​ D) Old row values are restored on rollback​

Answer: C) New rows appear in re-executed queries​

Explanation: Phantom reads are inserts by other transactions causing different query results.

87.

Which is true about strict two-phase locking?

●​ A) All locks are released after growing phase​

●​ B) Read and write locks can be released separately​

●​ C) No lock is released until commit or abort​

●​ D) Shared locks are always released first​

Answer: C) No lock is released until commit or abort​

Explanation: This ensures serializability and recoverability.

88.

A transaction is rolled back by the system. This is most likely due to:

●​ A) Deadlock prevention​
 ●​ B) Completion of execution​

●​ C) Log truncation​

●​ D) Phantom reads​

Answer: A) Deadlock prevention​

Explanation: To resolve deadlock, one of the transactions may be aborted and rolled back.

89.

Which ensures that a transaction reads the latest committed value?

●​ A) Read Uncommitted​

●​ B) Read Committed​

●​ C) Repeatable Read​

●​ D) Serializable​

Answer: B) Read Committed​

Explanation: It allows reading only committed data, preventing dirty reads.

90.

Which condition is not necessary for deadlock to occur?

●​ A) Mutual exclusion​

●​ B) Hold and wait​

●​ C) Preemption​

●​ D) Circular wait​

Answer: C) Preemption​
Explanation: Preemption is the absence of a condition required for deadlock. It's a solution,
not a cause.
91.

In timestamp ordering, what happens if TS(T) < RTS(X) during a write?

●​ A) Transaction writes​

●​ B) Transaction reads​

●​ C) Transaction is aborted​

●​ D) RTS(X) is updated​

Answer: C) Transaction is aborted​

Explanation: A younger transaction has already read the value, so older one cannot write.

92.

Which recovery method requires no undo?

●​ A) Deferred update​

●​ B) Immediate update​

●​ C) Shadow paging​

●​ D) Write-ahead logging​

Answer: C) Shadow paging​

Explanation: Old pages are preserved, and new pages replace only upon commit—no undo
needed.

93.

What is the role of log-based recovery?

●​ A) Tracking transaction order​

 ●​ B) Tracking user sessions​

●​ C) Recording changes for recovery​

●​ D) Managing locks and timestamps​

Answer: C) Recording changes for recovery​

Explanation: Logs store transaction actions for rollback or redo in case of crash.

94.

A transaction reads X and writes to Y. Another writes X. What schedule type could this lead to?

●​ A) View equivalent​

●​ B) Cascading rollback​

●​ C) Phantom anomaly​

●​ D) Deadlock​

Answer: B) Cascading rollback​

Explanation: If T1 reads uncommitted X from T2 and T2 aborts, T1 must rollback too.

95.

Which lock mode combination on same item leads to a conflict?

●​ A) S & S​

●​ B) X & X​

●​ C) S & Intent-S​

●​ D) S & X​

Answer: D) S & X​
Explanation: A shared and exclusive lock on the same item causes a conflict, blocking
access.
96.

In serial schedules, which is true?

●​ A) They always result in higher throughput​

●​ B) They reduce concurrency​

●​ C) They allow dirty reads​

●​ D) They allow deadlocks​

Answer: B) They reduce concurrency​

Explanation: Serial schedules execute one transaction at a time, reducing concurrency but
avoiding anomalies.

97.

Which anomaly is avoided by Repeatable Read but not by Read Committed?

●​ A) Dirty Read​

●​ B) Non-repeatable Read​

●​ C) Phantom Read​

●​ D) Lost Update​

Answer: B) Non-repeatable Read​

Explanation: Repeatable Read prevents changes to rows already read, but not phantoms.

98.

What’s the first action after system crash recovery?

●​ A) Clear buffer pool​

 ●​ B) Restore from last checkpoint​

●​ C) Reboot log manager​

●​ D) Abort all transactions​

Answer: B) Restore from last checkpoint​

Explanation: The system starts log scanning from the last checkpoint for efficient recovery.

99.

Which of these reduces log scanning time during recovery?

●​ A) Log compression​

●​ B) Checkpointing​

●​ C) Buffer flushing​

●​ D) Page swapping​

Answer: B) Checkpointing​
Explanation: Checkpoints provide a known good state and reduce the amount of log to
process.

100.

Which is not guaranteed by Read Committed isolation level?

●​ A) No dirty reads​

●​ B) No lost updates​

●​ C) No non-repeatable reads​

●​ D) Greater concurrency​
Answer: C) No non-repeatable reads​
Explanation: Read Committed allows different values on re-read of same item, hence
non-repeatable reads are possible.