The Idea
In a distributed database, the same data may live on several nodes. With eventual consistency, a write is accepted on one node and propagated in the background. For a short window, different replicas can return different values, but if writes stop, all replicas eventually converge to the same state.
Why Systems Choose It
- It keeps the system available even when nodes cannot talk to each other.
- Writes do not wait for every replica to acknowledge, so latency stays low.
- It trades immediate correctness for availability and performance.
What You Must Handle
- A read right after a write might return a stale value.
- Conflicting writes need a rule such as last write wins or a merge function.
- Apps that need a guarantee can request read your own writes behavior.
Key idea
Eventual consistency lets replicas accept writes independently and converge over time, trading temporary staleness for higher availability and lower write latency.