What Is Uptime?
Uptime is the percentage of time a system, service, or server is operational and available. It is a key metric in Service Level Agreements (SLAs) that hosting providers, cloud services, and IT departments commit to maintaining. The "nines" notation is a common shorthand: 99.9% is "three nines," 99.99% is "four nines," etc.
Higher uptime percentages exponentially reduce the allowed downtime. Going from 99% to 99.9% cuts allowed downtime by 10x. Going from 99.9% to 99.99% cuts it by another 10x. Achieving five nines (99.999%) requires significant investment in redundancy, failover systems, and monitoring.
Uptime Formula
Common SLA Uptime Levels
| Uptime % | Name | Downtime/Year | Downtime/Month |
|---|---|---|---|
| 99% | Two nines | 3d 15h 36m | 7h 18m |
| 99.5% | Two and a half nines | 1d 19h 48m | 3h 39m |
| 99.9% | Three nines | 8h 45m 36s | 43m 48s |
| 99.95% | Three and a half nines | 4h 22m 48s | 21m 54s |
| 99.99% | Four nines | 52m 33.6s | 4m 23s |
| 99.999% | Five nines | 5m 15.4s | 26.3s |
Frequently Asked Questions
What uptime should I expect from my hosting provider?
Most reputable hosting providers guarantee 99.9% uptime (about 8.76 hours downtime per year). Premium and enterprise providers offer 99.95% or 99.99%. Be sure to check what constitutes "downtime" in the SLA -- some exclude scheduled maintenance windows.
Does 99.9% uptime include maintenance?
It depends on the provider's SLA. Some providers exclude scheduled maintenance windows from their uptime calculation. Others include all downtime. Always read the fine print of the SLA to understand what is and is not covered.
How is uptime typically measured?
Uptime is measured using monitoring services that ping or request the service at regular intervals (typically every 1-5 minutes). If the service fails to respond, it is recorded as downtime. Common tools include Pingdom, UptimeRobot, StatusCake, and Datadog.