Gossip Protocol
maang.io System Design Series
A gossip protocol spreads information the way a rumor spreads through a crowd: each node periodically picks a few random peers and swaps state. In a handful of rounds — O(log N) — the whole cluster converges, with no central coordinator and no single point of failure. It's how Cassandra, DynamoDB, and Consul track membership and detect failures.
This deep dive comes in three parts:
- Foundations & Core Concepts — epidemic dissemination, push/pull, and anti-entropy.
- Internals — convergence math, failure detection (the phi-accrual detector), and membership, with worked numbers.
- Interview — Staff/Principal Q&A: convergence guarantees and where gossip fits vs consensus.
Gossip Protocol — Part 3: System Design & Interview Mastery
maang.io System Design Series
What is this?
You now know gossip from the office rumor down to the phi-accrual detector. This chapter turns that into interview points. Interviewers rarely ask "explain gossip" as a standalone quiz — instead, gossip shows up as the right answer to a sub-problem inside a bigger design: "how do your 500 nodes know who's alive?", "how does a new node join without a master?", "how do you detect a dead node without a central health-checker?" The senior move is to recognize that sub-problem and name gossip for it — and, just as important, to know its edges so you don't reach for it when you actually need consensus.
We'll do two things: see where gossip fits in real designs (with diagrams), then run a mock interview — for each question, what they're really asking, a model answer, and the senior line that signals you've operated this for real.
The one-line idea: interviewers use gossip to test whether you can build a coordinator-free, self-healing cluster — spread membership and detect failures among many peers with no single point of failure — and whether you know the exact boundary where "eventually consistent gossip" must hand off to "strongly consistent consensus."
Sign in to continue reading
The rest of this lesson is available with a free account. Signing in with Google or Microsoft is free.
Sign in to read the full lesson