Uptime & SLA downtime calculator
What does "99.9% uptime" actually allow? Convert any availability target into the downtime it permits — per day, week, month and year — and back again. Plus your error budget.
Your availability target
Each extra nine roughly divides the allowed downtime by ten. Pick the target the business needs, then architect for it.
At 99.9%, your 30-day error budget is the total downtime — across all incidents — you can spend before breaching the target.
Availability math assumes downtime is measured over the stated window; real SLAs define their own measurement windows, exclusions and service credits differently.
Downtime → availability
Had an outage this month? Enter the total minutes down to see the availability percentage it implies.
Uses 43,829.06 minutes per average month (365.25 × 24 × 60 ÷ 12). Availability = (1 − downtime ÷ 43,829.06) × 100.
The common nines
| Availability | Per day | Per week | Per month | Per year |
|---|
Computed live from the same formulas, so it stays consistent with the calculator above.
Picking a number is easy. Hitting it is engineering.
Redundancy, graceful degradation, timeouts, retries, health checks and an honest error budget — designed in before the incident, not during it. Book a free build audit and we'll pressure-test where your real availability stands.
Book a Build AuditReading an SLA in plain numbers
An availability percentage only means something once you convert it into time. "Three nines" sounds robust until you see it allows roughly 43 minutes of downtime a month — about the length of one bad deploy. This uptime calculator turns any target into allowed downtime per day, week, month and year, and the reverse calculator turns an incident's minutes back into the availability percentage it produced. The error budget figure is the same number framed as a spending limit: it's how much unreliability your target lets you spend before you breach it.
Treat the budget as a decision tool. While it's healthy, ship fast and take calculated risks; when it's nearly gone, freeze risky changes and pay down reliability debt. The table below is the back-of-envelope reference engineers carry in their heads — but the calculator above computes every figure dynamically so you can model the exact target you're being asked to sign.
| Nines | Availability | Downtime / year |
|---|---|---|
| Two | 99% | ~3.65 days |
| Three | 99.9% | ~8.77 hours |
| Three & a half | 99.95% | ~4.38 hours |
| Four | 99.99% | ~52.6 minutes |
| Five | 99.999% | ~5.26 minutes |
Questions about uptime & SLAs
How much downtime does 99.9% uptime allow?
99.9% (three nines) allows about 8 hours 46 minutes of downtime per year, roughly 43 minutes per month, or about 1 minute 26 seconds per day. Each additional nine cuts the allowed downtime by roughly 10x — 99.99% is about 52 minutes per year.
What is an error budget?
An error budget is the amount of downtime or failed requests your availability target permits in a window — for 99.9% over 30 days that's about 43 minutes. Teams spend it deliberately: while budget remains you can ship faster and take risks; when it's exhausted you freeze risky changes and focus on reliability.
How many nines do I actually need?
More nines cost exponentially more in redundancy, testing and on-call. Most products are well served by 99.9%; payment, infrastructure and life-critical systems justify 99.99%+. Pick the target the business actually needs, then architect for it — don't pay for nines your users won't notice.