SRE Reliability Budget Calculator
An SRE calculator that translates an availability objective into allowable downtime, measures error-budget consumption and indicates whether the current burn supports normal release activity.
Who this Lab is for
Designed for
- SRE and platform teams
- Service owners defining availability targets
- Engineering leaders balancing reliability and delivery
Use it when
- Defining or reviewing an availability SLO
- Assessing incident impact during an SLO period
- Deciding whether reliability work should take priority over releases
A complete run, step by step
Enter the SLO
Use the service's agreed availability target rather than an aspirational number.
Set the measurement window
Choose the same period used by your SLO reporting, commonly 28 or 30 days.
Add observed downtime
Enter unavailable minutes measured so far and the number of elapsed days.
Interpret burn
Review remaining budget and projected burn before making release or reliability decisions.
What you will need
Prepare the following information before starting. Use measured evidence where possible; defaults are examples and should not be treated as recommendations.
Availability SLO
Target availability for the period.
Budget period
Usually 28 or 30 days.
Observed downtime
Unavailability consumed so far.
Period elapsed
Used to calculate projected burn.
What the result tells you
Your report includes
- Downtime allowance for the selected period
- Consumed and remaining error budget
- Burn-rate and release guidance
How it is determined
Allowance equals total period minutes multiplied by the unavailable fraction of the SLO. The Lab compares observed downtime with elapsed time to estimate consumption and projected burn across the full period.
The arithmetic is deterministic, but downtime minutes approximate availability and may not match a request-based SLI.
Model assumptions
- • The SLO applies uniformly across the selected period.
- • Observed downtime uses the same SLI definition as the SLO.
- • The current consumption rate is projected from elapsed time.
Authoritative references
99.9% monthly SLO
Situation
A 30-day period allows roughly 43.2 unavailable minutes. The service records 18 minutes in the first 15 days.
Result
About 42% of the budget is consumed halfway through the period, indicating the service is currently burning slightly below the full-period allowance.
Use the result with engineering judgement
- Downtime minutes are a simplified proxy for request-based SLIs.
- The calculator does not define which events should count as unavailable.
- Use multi-window burn-rate alerting for operational paging.
Questions before you begin
Is 100% availability a useful SLO?
Usually not. A realistic target preserves room for change and focuses investment on reliability users actually need.
Should planned maintenance count?
That depends on the published SLO. Define exclusions explicitly and apply them consistently.
What happens when the budget is exhausted?
Teams commonly restrict risky changes and prioritise reliability work, but the exact policy should be agreed before the period begins.
Reliability Budget is under review
This legacy judgement-based Lab has been retired. Existing saved reports remain available, but new execution is disabled.
Open deterministic utilities