SLA Uptime Calculator: Downtime & Error Budget
See how much downtime an availability target allows per day, week, month and year, how much error budget you have left, and the burn-rate thresholds to alert on.
Free, no signup. Runs entirely in your browser; nothing you enter is sent anywhere.
Error budget (30-day window)
Budget 43m 12s, used 10m. Remaining: 33m 12s (76.9%).
Burn-rate alerts
| Severity | Windows | Burn rate | Alert if errors > |
|---|---|---|---|
| page | 1h & 5m | 14.4× | 1.44% |
| page | 6h & 30m | 6× | 0.6% |
| ticket | 3d & 6h | 1× | 0.1% |
Both windows must exceed the threshold. Thresholds follow the Google SRE Workbook for a 30-day SLO.
What the nines actually allow
Each extra nine cuts allowed downtime by ten. 99.9% allows about 43 minutes a month; 99.99% allows about 4 minutes, which is less time than most teams need to notice an incident, let alone fix it. Pick a target from what users actually need and what your dependencies can deliver: a service that calls two 99.9% dependencies in series can't promise more than about 99.8% itself.
Error budgets and burn-rate alerts
The error budget is the downtime (or failed requests) your target permits over a window, here 30 days. Spending it on planned risk such as deploys and migrations is fine; running out means slowing down. Burn rate is how fast you're spending it: a burn rate of 1 uses exactly the whole budget in 30 days. The table uses the multiwindow alerts from the Google SRE Workbook, which page on fast burns and open tickets for slow ones while ignoring short blips. Our guide to burn-rate alerting shows how to wire these into Prometheus.
Questions people ask
How much downtime is 99.9% uptime?
About 8 hours 46 minutes per year, 43 minutes 50 seconds per month, 10 minutes 5 seconds per week, or 1 minute 26 seconds per day. Month figures here use the average month length of 30.44 days.
How much downtime does 99.99% allow?
About 52 minutes 36 seconds per year, or roughly 4 minutes 23 seconds per month. At that level, recovery has to be automated, because a human-paced response usually takes longer than the whole monthly budget.
What is an error budget burn rate?
The ratio between how fast you are consuming your error budget and the rate that would use it up exactly at the end of the SLO window. A burn rate of 14.4 sustained for one hour consumes 2% of a 30-day budget, which is the usual threshold for paging someone.