Error budget calculator
Enter your SLO and what you have burned so far to see the budget in downtime minutes and failed requests, how much is left, your burn rate, and the multi-window burn-rate thresholds to alert on.
- Budget
- 43.2 min
- 50,000 failed requests
- Consumed
- 27.8%
- 31.2 min left
- Burn rate
- 0.83x
- on track
You are within budget at the current pace.
| Alert | Budget burned | Over | Burn rate |
|---|---|---|---|
| Page: fast burn | 2% | 1 h | 14.4x |
| Page: slow burn | 5% | 6 h | 6.0x |
| Ticket: steady leak | 10% | 3 days | 1.0x |
Everything runs in your browser; nothing you type is sent anywhere.
How to use it
- Set the SLO target and window, normally 99.9% over 30 days.
- Enter your request volume for the window, plus minutes of downtime and failed requests so far, and how many days have elapsed.
- Use the remaining budget and burn rate to decide whether to ship or stabilize, and copy the alert thresholds.
Frequently asked questions
What is an error budget?
The amount of unreliability your SLO allows. A 99.9% SLO over 30 days allows 43.2 minutes of downtime, or one failed request in a thousand.
What is a burn rate?
How fast you are spending the budget compared with an even spend. A burn rate of 1 uses exactly the whole budget by the end of the window, and 10 uses it in a tenth of the time.
Which burn-rate alerts should I set?
Google's SRE workbook suggests paging at 14.4x over 1 hour (2% of a 30-day budget), paging at 6x over 6 hours (5%), and opening a ticket at 1x over 3 days (10%).
What should happen when the budget runs out?
Agree the policy beforehand: typically freeze risky releases and spend engineering time on reliability until the budget recovers.
Made by compiler.dev, faster GitHub Actions runners. More free tools.
Related tools
- Uptime and SLA calculatorConvert uptime percentage (99.9%, 99.99%, five nines) into allowed downtime per day, week, month and year, and see how dependencies in series lower it.
- DORA metrics calculatorCalculate the four DORA metrics from your deploy data and see if you rank Elite, High, Medium or Low on each, using the State of DevOps bands.
- On-call rotation generatorGenerate an on-call rotation schedule from a list of people, a time zone and a shift length. See the calendar table and download an ICS file.
- Cost of slow CI calculatorWork out what slow CI costs your team each year from engineers, waits per day, minutes per wait and hourly cost, plus the saving from faster builds.