SLA Uptime Calculator
Enter an SLA percentage and see exactly how much downtime it allows per day, week, month, quarter or year — or work backwards from observed downtime to find your achieved SLA. All calculations run in your browser.
This means the service must be up at least — of the selected period, which is — uptime.
Common SLA reference table
Allowed downtime for the most common SLAs, per month (730 hours) and per year (8760 hours).
| SLA | Downtime / month | Downtime / year | Error budget share |
|---|
What is an error budget?
An SLA (service level agreement) promises a certain percentage of uptime, for example 99.9%. The error budget is the small fraction of time the service is allowed to be down: 100% minus the SLA.
Teams use the error budget to balance speed and reliability. If the service stays well within its budget, the team can ship new features and take risks. If the budget is nearly spent, the team freezes risky changes and focuses on reliability until the budget recovers.
Tip: a stricter SLA is much more expensive to meet. Moving from 99.9% to 99.99% cuts allowed downtime by 10x (from about 43 minutes to about 4 minutes per month), which is why most non-critical services pick 99.9%.
How to use
- Keep the "SLA to downtime" tab, enter an SLA percentage and a period, then click Calculate downtime.
- Or switch to "Downtime to SLA", enter the downtime you observed, and see your achieved SLA.
- Use the reference table to compare common SLA levels.
Note: A month is treated as 730 hours (365 days / 12) and a year as 8760 hours. Real calendar months vary by a day, so these are standard planning figures, not exact calendar values.