How to use this calculator
- Enter or choose uptime target.
- Enter or choose period.
- The allowed downtime updates as you type. Use Copy result or Share to keep it.
Method
Allowed downtime = Period length × (1 − Uptime% ÷ 100)
- Period length
- the measurement window in minutes (30 days = 43,200 min; 365 days = 525,600 min)
- Uptime%
- the availability target
Worked example
- A 30-day month is 30 × 24 × 60 = 43,200 minutes.
- Downtime share: 1 − 0.999 = 0.001.
- 43,200 × 0.001 = 43.2 minutes = 43 min 12 s.
A 99.9% uptime target allows 43 minutes 12 seconds of downtime in a 30-day month (8 h 45 min 36 s a year).
What the result means
Each extra nine cuts the allowance tenfold: 99.99% leaves about 4.3 minutes a month, which is less than most manual incident responses take. Higher targets need automated failover rather than faster humans.
Assumptions and limits
- Downtime is measured as whole-service unavailability over the chosen calendar period.
- Real SLAs define what counts as downtime (for example, excluding planned maintenance, or measuring in 5-minute intervals); read your contract’s definition.
Questions
Which month length should I use?
Use whatever your SLA uses. Many use a calendar month, so 28 to 31 days; 30 days and the average month (30.44 days) are common approximations.
Is availability the same as uptime?
Usually yes in SLAs. Request-based SLOs instead count the share of successful requests; use the Error Budget Calculator for those.