Why Replicate Data?
We just talked about storing data across multiple servers—horizontal scaling, partitioning. Now we have multiple servers, each holding a subset or a copy of the data. That already improves scale. But we also want redundancy: if one server dies, we should still be able to access the data from another. Database replication is how we get that. We duplicate data across multiple servers so we have backups and avoid a single point of failure.
What is database replication?
Replication means keeping copies of the same data on more than one server. If one database (or the server it runs on) crashes or is lost, we have a copy somewhere else. No single machine is the only place the data lives. So we can keep reading and serving traffic from the remaining replicas. That gives us redundancy and high availability—no single point of failure.