ACID properties in HLD - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Which ACID property ensures that a transaction is completed fully or not at all?

In database systems, transactions must be reliable. Which ACID property guarantees that either all steps of a transaction happen or none do, preventing partial updates?

A Durability

B Atomicity

C Consistency

D Isolation

💡 Hint

Think about the property that treats the transaction as a single unit.

Atomicity means the transaction is all or nothing. If any part fails, the whole transaction is rolled back.

❓ Architecture

intermediate

Which ACID property prevents transactions from interfering with each other?

When multiple users access a database at the same time, which ACID property ensures their transactions do not affect each other's results?

A Atomicity

B Durability

C Isolation

D Consistency

💡 Hint

Think about how transactions are kept separate during execution.

Isolation ensures that concurrent transactions do not see each other's intermediate states, preventing conflicts.

❓ scaling

advanced

In a distributed database, which ACID property is hardest to maintain when scaling horizontally?

When a database is spread across many servers to handle more users, which ACID property becomes most challenging to guarantee?

A Durability

B Atomicity

C Isolation

D Consistency

💡 Hint

Consider the difficulty of keeping data the same everywhere at the same time.

Consistency is difficult in distributed systems because all nodes must agree on data state despite network delays or failures.

❓ tradeoff

advanced

Which ACID property can be relaxed to improve database performance but risks reading uncommitted data?

To speed up database operations, sometimes one ACID property is weakened. Which one is it, and what risk does this introduce?

A Isolation; risk of dirty reads

B Atomicity; risk of partial transactions

C Durability; risk of losing committed data

D Consistency; risk of invalid data

💡 Hint

Think about concurrency and seeing changes before they are finalized.

Relaxing Isolation can improve speed but allows dirty reads, where a transaction sees uncommitted changes from others.

❓ estimation

expert

Estimate the minimum number of replicas needed to ensure Durability in a distributed database with potential node failures.

You want to guarantee that committed data is never lost even if some servers fail. What is the minimum number of replicas you need to store the data safely?

A 3 replicas, to tolerate one server failure and maintain quorum

B 2 replicas, to tolerate one server failure

C 1 replica, because one copy is enough if the server is reliable

D 4 replicas, to tolerate two server failures

💡 Hint

Think about how many copies are needed to survive one failure and still agree on data.

Three replicas allow one failure while still having a majority (quorum) to confirm data durability.