Suggest an editImprove this articleRefine the answer for “How does replication help scale a database?”. Your changes go to moderation before they’re published.Approval requiredContentWhat you’re changing🇺🇸EN🇺🇦UAPreviewTitle (EN)Short answer (EN)Replication helps scale a database by **spreading the load** across several servers: the primary server (master) handles only write operations, while every read request goes to replicas (read replicas) - this reduces load on the master and lets more users be served at once. **Key point:** replication also improves availability (a replica can be promoted to the new primary if the master fails), enables geographic scaling (replicas placed closer to users in different regions), and a gradual move toward horizontal scaling.Shown above the full answer for quick recall.Answer (EN)ImageReplication helps scale a database by **spreading the load** across several servers. ### How it works Here's how it works: 1. **Offloading reads.** The primary server (master) handles only write operations, `INSERT`, `UPDATE`, `DELETE`. Every read request (`SELECT`) goes to replicas (read replicas). This reduces load on the master and lets more users be served at once. 2. **Improving availability.** If the master fails, a replica can be promoted to the new "primary" (failover). This makes the system more resilient and cuts downtime. 3. **Geographic scaling.** Replicas can be placed closer to users in different regions. This reduces read latency. 4. **Architectural flexibility.** Replication allows a gradual move toward horizontal scaling, adding new nodes without stopping the system. So replication is a key tool for scaling a database's reads, improving fault tolerance, and adding flexibility to the whole architecture.For the reviewerNote to the moderator (optional)Visible only to the moderator. Helps review go faster.