Skip to main content
Legia

IT Infrastructure & Cloud

What is Uptime?

Also called service availability, availability, downtime.

Short definition

Uptime is the percentage of time a system, such as a website, server, or application, is running and accessible over a given period. The opposite is called downtime.

Reading uptime figures

A year has 8,760 hours. 99% uptime means the system may be down for 1% of that time, about 87.6 hours or three and a half days per year. 99.9% uptime means a downtime allowance of 0.1%, about 8.76 hours per year, or about 44 minutes per month. 99.99% uptime leaves about 52.6 minutes per year. Each additional nine cuts the downtime allowance to a tenth, and usually raises infrastructure costs sharply.

What to ask

An uptime figure in a contract only means something if it is clear how it is measured. Is it measured per month or per year, where is it monitored from, does scheduled maintenance count, and what is the compensation if it is missed? A hosting provider's server uptime is also different from your application's uptime: the server can be running while the application is throwing errors.

Example in a company

An ordering application that customers use around the clock deserves a higher target than an internal reporting application that is only opened during work hours. To reach a high target, a company needs more than one server, a load balancer, automated monitoring, and a recovery procedure that has been rehearsed. A lower but honest target is more useful than a high figure that is never measured.

Related Legia service Server Installation & Management Setup, hardening, monitoring, and migration for physical, virtual, and cloud servers.

Resources