
Definition (in the context of systems and data centers):
High Availability (HA) is a set of designs and practices that ensure a computer system or service remains operational and accessible most of the time, minimizing interruptions and downtime.
Main characteristics:
- Redundancy: duplication of key resources (servers, networks, power) to avoid single points of failure.
- Continuous monitoring: ongoing supervision of components to proactively detect and solve problems.
- Automatic recovery: ability to transfer the load to alternative systems when a component fails.
Example:
A web service with high availability can continue to function without interruptions even if one of its servers fails, as other servers immediately take over the load.
Key objective:
Maintain a stable and reliable service level, often expressed as a percentage of uptime (for example, 99.9% annually).
See also: Redundancy, Fault Tolerance.
« Back to Glossary Index