1. What does the term "consistency model" refer to in a distributed system?
A. How strictly the database enforces foreign key constraints
B. Rules for how and when updates become visible to different nodes
C. The encryption method for securing data at rest
D. The speed at which data can be replicated
2. Which of the following best describes linearizability?
A. All writes must happen in the same physical location
B. Ensuring each operation appears to occur atomically and in a globally consistent order
C. Disabling concurrency in transactions
D. Allowing updates without any ordering guarantees
3. Why are consensus algorithms like Raft or Paxos crucial in distributed systems?
A. They ensure all data is stored on a single node for simplicity
B. They provide a method for nodes to agree on a single value or sequence of operations, despite failures
C. They eliminate the need for replication entirely
D. They ensure perfect network reliability at all times
4. What limitation does the CAP theorem place on distributed systems?
A. They can scale unlimitedly if the network never fails
B. They can only guarantee two out of three: consistency, availability, and partition tolerance
C. They must always sacrifice partition tolerance in favor of strict consistency
D. They can only use one type of data model to remain consistent