Error-budget example: request allowance versus bad minutes
Compare distinct request-based and time-based availability budgets with explicit denominators.
Illustrative dataset · Tool version 2 · Reviewed September 29, 2026
What does the budget actually count?
At 99.9%, 1,000,000 eligible events allow 1,000 bad events. With 250 bad events, 750 remain and 25% is consumed. For a separately defined 30-day time-based availability objective, 43,200 eligible minutes allow 43.2 bad minutes; 15 bad minutes leave 28.2. Switching definitions is not converting requests into downtime.
| Measure | Events |
|---|---|
| Allowed | 1000 |
| Bad | 250 |
| Remaining | 750 |
What this example does not establish
Exclusions and missing data change the denominator. A policy checkpoint records your own agreed rule and does not authorize a release.
How a different input changes the result
Double both the eligible and bad-event counts while keeping the 99.9% target. The remaining allowance doubles, while consumption remains 25%. This distinguishes a percentage of budget used from the absolute number of failures still allowed.
Read the related guide · Source and further reading
These datasets and worksheets may be adapted for your team or training. Attribute AIOpsSRE and link to this example when redistributing. Third-party sources retain their own terms.
Get AIOpsSRE by email · Report a calculation or explanation issue