Turn an SLO into a concrete error budget in minutes and failed requests, and track the burn rate against it.
An SLO of 99.9% over 30 days is 43.2 minutes of allowed failure, and the same fraction applied to request volume gives the number of failed requests that fit inside the objective. The burn rate normalises consumption against elapsed time so that 1.0 is on-pace regardless of where you are in the period. Error budgets convert reliability arguments into arithmetic: while budget remains you can ship changes, and when it is gone the answer to a risky change request is no.
Network Error Budget
error budget = (1 − SLO) × period; allowed failures = (1 − SLO) × requests; burn rate = share of budget spent ÷ share of period elapsed.
error budget = (1 − SLO) × period; allowed failures = (1 − SLO) × requests; burn rate = share of budget spent ÷ share of period elapsed. An SLO of 99.9% over 30 days is 43.2 minutes of allowed failure, and the same fraction applied to request volume gives the number of failed requests that fit inside the objective. The burn rate normalises consumption against elapsed time so that 1.0 is on-pace regardless of where you are in the period.
Error budgets convert reliability arguments into arithmetic: while budget remains you can ship changes, and when it is gone the answer to a risky change request is no.
This calculator takes 5 inputs: Service level objective, Budget period, Requests in the period, Budget consumed so far, Period elapsed. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.