Suggest an editImprove this articleRefine the answer for “What does "master-replica" architecture mean?”. Your changes go to moderation before they’re published.Approval requiredContentWhat you’re changing🇺🇸EN🇺🇦UAPreviewTitle (EN)Short answer (EN)The **"master-replica" architecture** (or "primary-replica") is a replication model where one database server is the **primary (master)**, and the rest are **copies (replicas / slaves)**: every write operation (`INSERT`, `UPDATE`, `DELETE`) runs **only on the master server**, which then passes changes to the replicas in real time, and those handle the read operations (`SELECT`). **Key point:** this architecture reduces load on the master, improves fault tolerance (a replica can be promoted to the new primary), and enables read scaling, but it requires data-consistency mechanisms and failover processes.Shown above the full answer for quick recall.Answer (EN)ImageThe **"master-replica" architecture** (or "primary-replica") is a replication model where one database server is the **primary (master)**, and the rest are **copies (replicas / slaves)**. ### How it works 1. Every **write operation** (`INSERT`, `UPDATE`, `DELETE`) runs **only on the master server**. 2. The master **passes changes** to the replicas in real time through the replication mechanism. 3. The replicas hold up-to-date copies of the data and handle **read operations** (`SELECT`). ### Why this matters - it reduces load on the master, since reads get spread across replicas; - it improves fault tolerance, if the master fails, a replica can be promoted to the new primary; - it enables read scaling and speeds up the application. Important: this architecture requires **data-consistency mechanisms** (so replicas don't fall behind) and **failover processes** (automatic switchover to a new primary node on failure).For the reviewerNote to the moderator (optional)Visible only to the moderator. Helps review go faster.