Suggest an editImprove this articleRefine the answer for “What is horizontal scaling (horizontal scaling)?”. Your changes go to moderation before they’re published.Approval requiredContentWhat you’re changing🇺🇸EN🇺🇦UAPreviewTitle (EN)Short answer (EN)**Horizontal scaling** is increasing a system's performance by adding new servers (nodes) that jointly handle the database's load: instead of one powerful server, a **cluster** of several machines is used, with data and requests split across them via sharding, replication, and load balancing. **Key point:** the advantage is nearly unlimited performance growth and fault tolerance; the drawback is added complexity in data synchronization, transaction consistency, and cluster administration.Shown above the full answer for quick recall.Answer (EN)Image**Horizontal scaling** is increasing a system's performance by adding new servers (nodes) that jointly handle the database's load. ### The idea is simple Instead of one powerful server, a **cluster** of several machines is used. Data and requests get split across them. The main methods: - **Sharding**, splitting data into parts, each stored on a separate server. For example, users with IDs 1-100000 on one server, 100001-200000 on another. - **Replication**, copying data between nodes. One server can be the primary (master), while the rest are copies (replicas) that only accept read requests. - **Load balancing**, spreading requests across servers for an even load. The advantage is nearly unlimited performance growth and fault tolerance. The drawback is added complexity in data synchronization, transaction consistency, and cluster administration.For the reviewerNote to the moderator (optional)Visible only to the moderator. Helps review go faster.