Distributed Systems Practice Questions: Test Your Knowledge | LearnByTeaching.ai
Challenge your understanding of distributed systems with these 40 questions covering consensus, replication, consistency models, and fault tolerance. These questions test concepts from Lamport's foundational work through modern cloud-native architectures, at a level appropriate for graduate students and senior engineers.
40 questions total
Consensus and Leader Election
Test your understanding of consensus and leader election.
In the Raft consensus algorithm, what is the minimum number of nodes that must agree for a log entry to be committed in a 5-node cluster?
What problem does the Paxos 'prepare' phase solve that a naive 'just propose a value' approach would not?
In a Raft cluster, a leader becomes network-partitioned from the majority but can still communicate with two followers. What happens?
Which of the following is NOT a property guaranteed by the FLP impossibility result?
What is the primary advantage of Multi-Paxos over basic Paxos?
In leader election, what is the purpose of randomized election timeouts in Raft?
What is the Byzantine Generals Problem fundamentally about?
A ZooKeeper ensemble of 7 nodes experiences the failure of 3 nodes. Can the system still process write requests?
What does the 'term' number in Raft represent?
In PBFT (Practical Byzantine Fault Tolerance), why are three phases (pre-prepare, prepare, commit) needed instead of two?
Replication and Consistency Models
Test your understanding of replication and consistency models.
Which consistency model guarantees that if process A completes a write before process B starts a read, B will see A's write?
What is the key difference between linearizability and serializability?
In a system using quorum-based replication with N=5 replicas, W=3 write quorum, and R=3 read quorum, is strong consistency guaranteed?
What is the main purpose of vector clocks in distributed systems?
In Amazon's Dynamo, what technique is used to resolve write conflicts when concurrent updates occur?
What does 'read-your-writes' consistency guarantee?
Why might a system choose eventual consistency over strong consistency?
In chain replication, how do reads and writes flow?
What is a CRDTs (Conflict-free Replicated Data Type) main advantage in distributed systems?
What is the difference between primary-backup replication and multi-leader replication?
Fault Tolerance and Partitioning
Test your understanding of fault tolerance and partitioning.
According to the CAP theorem, which property must be sacrificed during a network partition?
What is a split-brain scenario in distributed systems?
What is consistent hashing primarily used for in distributed systems?
In Google's Spanner, how does TrueTime enable externally consistent transactions?
What is a fencing token and why is it used?
What does the PACELC theorem add to the CAP theorem?
In a distributed system, what is a 'gray failure'?
What is the purpose of anti-entropy protocols like Merkle trees in distributed storage?
How does the sidecar pattern improve fault tolerance in microservice architectures?
What is the fundamental problem with using wall-clock timestamps for ordering events in a distributed system?
Distributed Transactions and Data Processing
Test your understanding of distributed transactions and data processing.
In two-phase commit (2PC), what happens if the coordinator crashes after sending 'prepare' but before sending 'commit' or 'abort'?
What is the Saga pattern in distributed systems?
In the MapReduce programming model, what is the role of the shuffle phase between Map and Reduce?
What advantage does event sourcing provide over traditional CRUD in distributed systems?
How does three-phase commit (3PC) improve upon two-phase commit (2PC)?
In stream processing, what is the difference between event time and processing time?
What is exactly-once semantics in stream processing, and how is it typically achieved?
What problem does the outbox pattern solve in microservice architectures?
In Google's MapReduce, what is the purpose of the 'backup task' mechanism?
What is a distributed snapshot, and which algorithm is most commonly associated with it?
Scoring Guide
Total possible: 40
Study Recommendations
- Read Designing Data-Intensive Applications by Martin Kleppmann cover to cover — it is the single best resource for building distributed systems intuition
- Study the Raft consensus paper in detail and implement a basic version to solidify your understanding
- Review Jepsen test reports to understand how real distributed databases fail under network partitions
- Practice system design interview problems that require reasoning about consistency, availability, and partition tolerance tradeoffs
- Work through MIT 6.824 Distributed Systems labs for hands-on experience with real distributed protocols
More Distributed Systems Resources
Want more distributed systems practice?
Upload your notes and LearnByTeaching.ai generates unlimited practice questions tailored to your course. Then teach the concepts to AI students who challenge your understanding.
Try LearnByTeaching.ai — It's Free