# High availability

> Designing a system so it keeps working when one of its parts fails.

- Level: Intermediate
- Topics: Cloud
- URL: https://tidbit.legios.com.ar/high-availability/
- In Spanish: [Alta disponibilidad](https://tidbit.legios.com.ar/es/high-availability/)

**High availability** (HA) is the goal of keeping a service up nearly all the time, even when individual parts break. Servers crash, disks fail, networks drop and whole data centers lose power; a highly available system is built so that none of those events, on its own, takes the service down. It is usually measured as a percentage of uptime, like 99.9% or 99.99%.

The core idea is **redundancy**: no **single point of failure**. Instead of one server, run several behind a load balancer. Instead of one data center, spread them across two or three **availability zones**, separate buildings in the same cloud region. Keep a standby copy of the database that takes over automatically if the main one fails, a switch called **failover**. Health checks constantly test each piece and route traffic away from anything that stops answering.

High availability is about everyday failures and keeps the service running through them, often without anyone noticing. Disaster recovery is the plan for the rare big event that knocks out a whole region. Every extra "nine" of uptime costs more in duplicated machines and engineering effort, so teams pick a target that matches how much an outage really hurts, and write it down in an SLA.

## An analogy

A plane with two engines. It can fly on one if the other fails, so a single engine problem is an incident, not a crash. You pay for the second engine on every flight so that you never need to think about it.

## Keep in mind

- Keeps a service up when individual servers, disks or data centers fail.
- Built on redundancy: no single point of failure, spread across availability zones.
- Health checks and automatic failover move traffic away from broken parts.
- Each extra nine of uptime costs more; match the target to the real impact.

## Go deeper

- [Reliability pillar, AWS Well-Architected](https://docs.aws.amazon.com/wellarchitected/latest/reliability-pillar/welcome.html)
- [High availability on Wikipedia](https://en.wikipedia.org/wiki/High_availability)

## Related terms

- [Server](https://tidbit.legios.com.ar/server/): A computer whose job is to answer requests from other computers, all day, every day.
- [Data center](https://tidbit.legios.com.ar/data-center/): A building full of servers where the internet, the cloud and AI actually run.
- [SLA](https://tidbit.legios.com.ar/sla/): A written promise about how reliable a service will be, and what happens if it is not.
- [Load balancer](https://tidbit.legios.com.ar/load-balancer/): The traffic cop that spreads incoming requests across many servers.
- [Regions and Availability Zones](https://tidbit.legios.com.ar/regions-and-availability-zones/): Where AWS physically lives: cities worldwide, each split into separate data centers.
- [Cloud computing](https://tidbit.legios.com.ar/cloud-computing/): Renting computers over the internet instead of buying your own.
- [Database](https://tidbit.legios.com.ar/database/): An organized store of data that programs can search and update safely.
- [Disaster recovery](https://tidbit.legios.com.ar/disaster-recovery/): The plan for getting systems and data back after a major failure.