CS Free MCQ Bank

[ Option D ]

Two-Phase Locking (2PL) protocol, which is used for concurrency control in database systems, has the problems of deadlock and cascading rollback.

• Deadlocks occur because transactions may wait indefinitely for locks held by other transactions, causing a cycle of waiting and thus freezing progress.
• Cascading Rollback happens if a transaction fails and rolls back, causing other transactions dependent on its uncommitted data to also roll back, leading to wasted resources and inefficiency.

[ Option B ]

Concurrency Control in a database ensures that when multiple transactions run at the same time, they do not interfere with each other. Its main goal is to maintain correctness, consistency, and isolation so that the final result remains the same as if transactions were executed one by one.
When we check concurrency control protocols, we look for two important properties:

• Conflict Serializability: The schedule should behave like a correct serial order.
• Freedom from Deadlock: Transactions should not end up waiting for each other forever.

Two Phase Locking (2PL) ensures conflict serializability by enforcing that transactions lock data in two phases (growing and shrinking), but it does not guarantee freedom from deadlock. The deadlocks can still occur in 2PL.
Timestamp Ordering Protocol guarantees both conflict serializability and freedom from deadlock because it assigns timestamps to transactions and schedules operations based on these timestamps, avoiding cyclic waits and deadlocks.
Multiple Granularity Protocol primarily helps reduce lock overhead and increase concurrency by using locks at different levels of granularity, but it does not guarantee deadlock freedom by itself. The deadlock can still occur.

[ Option A ]

Two-Phase Locking (2PL) is a concurrency control protocol used in databases to ensure serializability of transactions.
Strict 2PL and Rigorous 2PL are two variants of the two-phase locking protocol. In strict 2PL, only exclusive locks are held until the transaction commits, while shared locks can be released earlier. 
In rigorous 2PL, all locks, both shared and exclusive are held until the end of the transaction. This makes rigorous 2PL stricter and ensures that no other transaction can access any locked data until the current transaction completes, providing serializability and strictness.

• Deadlocks may still occur in both protocols.

[ Option D ]

Locking is a concurrency control technique used in DBMS to ensure that multiple transactions do not interfere with each other.
Lost update occurs when two transactions update the same data and one update overwrites the other. Locking prevents this by allowing only one transaction to modify data at a time.
Uncommitted dependency or dirty read happens when a transaction reads data written by another uncommitted transaction. Locking avoids this by restricting access until commit.
Inconsistent data results from simultaneous uncontrolled access to data, which locking helps prevent.