DataDad.
Join the closed betaJoin beta
DataDad / Free tools

Uptime calculator

Enter an uptime target to see the downtime it allows per day, week, month, quarter, and year. Or enter real downtime to get your uptime percentage.

Uptime target

A month is 30.44 days, the average calendar month. A quarter is three average months. A year is 365 days.

Allowed downtime per month

43m 50s

Per day
1m 26s
Per week
10m 5s
Per month
43m 50s
Per quarter
2h 11m 29s
Per year
8h 45m 36s
Nines
3 nines

Calendar months: 28 days allow 40m 19s, 30 days allow 43m 12s, 31 days allow 44m 38s.

With a check every five minutes, 8 failed checks use up this month's budget. Each failed check counts as five minutes down.

A year at 99.9% uptime

8h 45m 36s of downtime in a year: 36.5% of one day.

UpPeriod with downtimeEach square is six hours. Any downtime makes the whole square red. Red squares do not measure outage duration. This illustration groups downtime at the start of the year; real outages occur at different times.

Uptime and downtime table

Allowed downtime for common uptime targets. Select a percentage to open it in the calculator.

Allowed downtime by uptime percentage, per day, week, 30.44-day month, and 365-day year
UptimeNinesPer dayPer weekPer monthPer year
1 nine2h 24m16h 48m3d 1h 2m 55s36d 12h
1.3 nines1h 12m8h 24m1d 12h 31m 27s18d 6h
1.6 nines28m 48s3h 21m 36s14h 36m 35s7d 7h 12m
2 nines14m 24s1h 40m 48s7h 18m 17s3d 15h 36m
2.3 nines7m 12s50m 24s3h 39m 9s1d 19h 48m
3 nines1m 26s10m 5s43m 50s8h 45m 36s
3.3 nines43.2s5m 2s21m 55s4h 22m 48s
4 nines8.6s1m4m 23s52m 34s
4.3 nines4.3s30.2s2m 11s26m 17s
5 nines864ms6s26.3s5m 15s
6 nines86ms605ms2.6s31.5s

Availability of services that work together

Your website is only up when the services it needs are up. Enter the availability of each one.

    Combined availability

    —

    Downtime per month
    —
    Downtime per year
    —

    This assumes the services fail independently. A shared cause, such as one data center or one bad deploy, makes the real result worse.

    How the calculator works

    Uptime is the share of time a service works. Downtime is the rest. Allowed downtime equals the period length multiplied by (100% − uptime).

    For 99.9% over a 365-day year: 365 × 24 × 60 × 60 seconds × 0.001 = 31,536 seconds, or 8h 45m 36s. For an average month of 30.44 days, 99.9% allows 43m 50s.

    Going the other way, uptime equals 1 − downtime ÷ period length. We cut off extra decimals instead of rounding. A rounded-up result could show a target as met when it was missed.

    Example: one outage and a 99.95% monthly target

    Your hosting agreement promises 99.95% uptime each calendar month. In a 31-day month, that allows 22m 19s of downtime.

    One 30-minute outage in that month gives 1 − 1,800 ÷ 2,678,400 = 99.932% uptime. The target is missed by about eight minutes.

    The same outage in a 365-day year still gives 99.994% for the year. The measurement period changes the answer.

    Common questions

    How much downtime does 99.9% uptime allow?

    99.9% uptime allows 1m 26s of downtime per day, 10m 5s per week, 43m 50s per average month, and 8h 45m 36s per 365-day year.

    What does “five nines” mean?

    Five nines means 99.999% uptime. It allows 5m 15s of downtime per year, or 26.3 seconds per average month. Each extra nine cuts the allowed downtime by a factor of ten.

    What counts as downtime?

    Your service level agreement (SLA) defines it. Common rules count failed requests, server errors, timeouts, and very slow responses. Many agreements leave out planned maintenance that you announce in advance. Some count an outage only if it lasts longer than a few minutes.

    What is an SLA, and how is it different from an SLO?

    An SLA is a promise to customers. It names an uptime target, a measurement period, and what happens if you miss it, often a service credit. An SLO is an internal target. Set your SLO stricter than your SLA so you notice problems before customers can claim credits.

    Which month and year length does this calculator use?

    A month is 30.44 days, the average calendar month (365.2425 ÷ 12). A year is 365 days. Calendar months are 28 to 31 days long, so the real budget changes each month. At 99.9%, a 28-day month allows 40m 19s and a 31-day month allows 44m 38s. A 366-day leap year allows 8h 47m 2s.

    How often should I check my website?

    Check at least as often as your downtime budget allows. A 99.99% monthly target allows 4m 23s. With a check every five minutes, one failed check already uses more than that. With 99.9%, the same checks can fail 8 times a month before you miss the target.

    How do dependencies change my uptime?

    If your website needs two services that each have 99.9% uptime, multiply them: 0.999 × 0.999 = 99.8%. That doubles the allowed downtime to about 1h 27m 37s per month. Redundant copies work the other way. Two independent copies at 99% are both down only 0.01% of the time, so together they reach 99.99%.

    Measure your real uptime.

    DataDad Uptime checks your website every five minutes. It records each outage and keeps a year of uptime history. Your first monitor is free.

    See how DataDad Uptime works

    No credit card required.

    Put the result to work

    Check a website now or start free uptime monitoring to measure actual availability.