Uptime SLA Calculator
Convert an availability percentage into allowed downtime per day, month and year.
Please note: An estimate based on the figures you entered. Real-world results vary with conditions this calculation cannot see, so treat it as a starting point rather than a final answer.
What the Uptime SLA Calculator does
An availability percentage translates directly into an allowance of downtime. The jump between tiers is steeper than it looks: going from 99.9% to 99.99% cuts the yearly budget from nearly nine hours to under an hour, and each further nine is another factor of ten.
Formula
Downtime = (1 − Availability) × Period lengthNines = −log₁₀(1 − Availability)Series availability = A₁ × A₂ × … × Aₙ
Inputs explained
| Input | Unit | Required | Notes |
|---|---|---|---|
| Start from | one of 2 options | Yes | — |
| Availability | % | In some modes | Accepts 0 or more, up to 100. Shown Start from is An availability target. |
| Downtime | minutes | In some modes | Accepts 0 or more. Shown Start from is An amount of downtime. |
| Downtime measured over | one of 4 options | In some modes | Shown Start from is An amount of downtime. |
| Components in series | number | Optional | Availability multiplies across dependencies, so more components means less overall uptime. Accepts 1 or more. |
How to use it
- Choose Start from.
- Fill in the remaining inputs the form shows for your choice.
- Optionally add Components in series.
- Select Calculate.
Worked example
A 99.9% availability target — three nines.
- Availability
- 99.9%
0.1% of a year is 8 h 45 min 36 s. That is 43 min 12 s a month, 10 min 5 s a week, or 1 min 26 s a day.
Reading the result
- "Cost of the extra components" (shown once components exceed 1) quantifies the downtime price of adding a dependency in the same time-per-year units as the headline figure, rather than as an abstract percentage drop.
- The precise downtime-per-year figure matters more than the "nines" label — two systems both loosely described as "three nines" can have meaningfully different actual downtime once the exact percentage is converted.
Assumptions and limitations
- Series-component math assumes every dependency fails independently — multiplying availabilities is only strictly correct when failures are not correlated, and shared infrastructure like a common power grid or cloud region can make combined availability worse than the simple product suggests.
- The calculation has no visibility into what actually counts as "downtime" under a given SLA — the notes flag that providers routinely exclude planned maintenance from that definition.
Common mistakes
- Assuming this series model applies to redundant components too — it models components in series, where any one failure takes the service down; redundant components in parallel, where either one suffices, would instead improve availability, the opposite effect, which this tool does not model.
Frequently asked questions
What does "five nines" actually mean?
99.999% availability — about 5 minutes 15 seconds of downtime a year. It demands full redundancy, automated failover and no single point of failure.
Why does adding components reduce availability?
Because in a series chain any one failure takes the whole service down, so the availabilities multiply. Five 99.9% components give 99.5%, not 99.9%.