Service-level agreements set a target availability—say 99.95 %—that allocates a precise error budget, the downtime you may consume without violating contractual reliability. Translating that percentage into seconds clarifies operational risk and guides engineering priorities.

The calculator combines your SLA target, reporting period, and any recorded downtime or planned maintenance. A reactive engine applies simple arithmetic to expose allowed downtime, remaining budget, and burn rate, then visualises usage in a lightweight charting layer.

Teams use the figures to schedule releases, justify change freezes, and alert stakeholders before breaching commitments. Values round to whole seconds; marginal discrepancies may appear at millisecond precision.

Technical Details:

An SLA expresses availability as a percentage, leaving 100 − SLA % as downtime. Multiplying that fraction by a period’s total seconds yields the maximum interruption window. The tool then sums incident minutes and scheduled maintenance to gauge consumption and remaining capacity.

Main Formula

D = P \times \frac{100 - S}{100}

D = allowed downtime (s); P = period length (s); S = SLA target %.

SLA % (30 days)	Allowed downtime
99.0 %	7 h 12 m 28 s
99.9 %	43 m 12 s
99.99 %	4 m 19 s
99.999 %	26 s

Smaller budgets force stricter change-control and observability; exceeding the limit breaks the SLA.

SLA target – desired availability percentage.
Period length – monitoring window in seconds, minutes, hours, days, weeks, or years.
Downtime experienced – unplanned interruption duration.
Planned maintenance – approved service windows counted against budget.

Example (99.9 % over 30 days, 50-minute outage):

30 \times 86 400 = 2 592 000

2 592 000 \times \frac{0.1}{100} = 2 592

50 \times 60 = 3 000

Budget exceeded by 408 s; burn = 115.73 %.

Linear scaling assumes independence between outages and period length.
No partial-credit for degraded performance — uptime is binary.
Inputs below one second are rounded.
Results guide planning; contractual interpretation may differ.

Periods of zero seconds yield undefined budgets.
SLA above 100 % resolves to negative downtime.
Maintenance outside the period is ignored.
Clock skew across distributed incidents may mis-align totals.

Practices align with the error-budget model popularised in the Google SRE handbook and elaborated by contemporary reliability research.

No personal or sensitive data is processed; calculations run locally in the browser.

Step-by-Step Guide:

Follow these steps to size, track, and export your error budget.

Enter a target availability in the SLA target field.
Specify the monitoring window under Period length.
(Optional) Add downtime experienced and planned maintenance so far.
Review allowed downtime, remaining budget, and burn percentage in the “Budget Summary” tab.
Switch to “Usage Chart” for a visual breakdown or “Allowed Downtime” to copy or download a CSV.

FAQ:

What counts as downtime?

Any period during which end-users cannot perform intended actions, including full outages and severe partial degradations.

Is maintenance always subtracted?

Only planned, announced windows agreed with stakeholders should be logged; unscheduled patches belong in downtime.

How precise are the figures?

Results round to whole seconds; micro-outages aggregate into the next full second.

Can budgets roll over?

No, each reporting period resets; unused allowance expires and cannot be banked for future windows.

Is my data stored?

All inputs stay in your browser session and vanish when the page closes; nothing is transmitted to any server.

Glossary:

SLA: Contractual availability target.
Error budget: Allowed downtime before breach.
Downtime: Unplanned service interruption.
Maintenance: Scheduled, approved outage.
Burn rate: Error-budget consumption speed.

Uptime Calculator

Allowed Downtime

Introduction:

Technical Details:

Step-by-Step Guide:

FAQ:

Glossary: