Definition

Horizontal scaling (scale-out) means adding more machines (nodes) to a system to handle increased load.

Core concept

Instead of upgrading one server, you add parallel machines. The workload and data are distributed among them. More nodes = more capacity linearly.

Use cases

Web servers and databases that partition (shard) data can scale horizontally almost indefinitely. Cloud-native applications leverage elastic horizontal scaling.

Trade-offs

Requires designing system to distribute data/requests (sharding, load balancing). It adds inter-node communication and consistency challenges.

Example

A MongoDB cluster with multiple shards can add new shard servers to store more data and serve queries.

References

Vertical vs horizontal: “Horizontal scaling – adding new nodes to manage the distributed workload”[[38]](https://www.freecodecamp.org/news/horizontal-vs-vertical-scaling-in-database/ "Horizontal vs. Vertical Scaling – How to Scale a Dat…").