Glossary
Mean time between failures (MTBF)
MTBF is mean time between failures: the average time a service runs between one failure and the next, a measure of how often it fails rather than how badly.
Also called MTBF.
MTBF stands for mean time between failures, and the MTBF formula is total operating time divided by the number of failures. Take a 720-hour month with 3 outages totalling 6 hours: the service ran for 714 hours, so MTBF is 714 / 3 = 238 hours. MTTR for the same month is 6 / 3 = 2 hours, and the two together give availability: 238 / (238 + 2) = 99.17%, the same figure as 714 hours up out of 720.
Mean time between failures is the average interval a service runs before the next failure. It is the frequency half of reliability, and it pairs with MTTR, which is the duration half. Together they produce availability: a service that fails rarely but takes a long time to recover and one that fails often but recovers instantly can post the same uptime number and be completely different things to live with.
The term arrives from hardware, where it describes the expected life of a physical component, and it transfers imperfectly. Software does not wear out; it fails when something changes — a deploy, a traffic pattern, a dependency, a certificate expiring. A software MTBF is therefore a description of a period that has passed rather than a prediction of one that has not, and it moves the moment the change rate does.
It is most useful as a trend on a single service, and least useful compared across services or quoted as a target. A falling MTBF says the system is becoming less stable, which is a real and early signal. A rising one may mean things improved, or may mean the team stopped shipping.
| Metric | Full form | What it measures |
|---|---|---|
| MTBF | Mean time between failures | How often a repairable system fails |
| MTTF | Mean time to failure | How long a part that is replaced, not repaired, lasts before it fails for good |
| MTTR | Mean time to recovery, or to repair | How long it takes to restore service once a failure starts |
| MTTD | Mean time to detect | How long a failure goes unnoticed |
Questions
What is MTBF?
MTBF, mean time between failures, is how often something fails. A service with an MTBF of 238 hours fails about once every ten days on average. It says nothing about how long each failure lasts; that is MTTR.
What is the full form of MTBF?
MTBF stands for mean time between failures. It is the average time a system runs between one failure and the next, and it is used for software services as well as for the hardware the term comes from.
How do you calculate MTBF?
Divide the total time the service was running by the number of failures in the same period. Over a 720-hour month with 3 outages totalling 6 hours, the service ran 714 hours, so MTBF is 714 / 3 = 238 hours. Leave the downtime out of the running time, or the figure flatters itself.
What is the difference between MTBF and MTTR?
MTBF measures how often a service fails, and MTTR measures how long each failure lasts. Availability comes from both: MTBF / (MTBF + MTTR). An MTBF of 238 hours with an MTTR of 2 hours gives 99.17%.
What is the difference between MTBF and MTTF?
MTBF applies to things that are repaired and put back into service, like a web service or a server. MTTF, mean time to failure, applies to things that are replaced rather than repaired, like a disk, and measures how long one lasts before it fails for good.
Uptimely records when every incident was declared and resolved, so how often a service fails and how long it stays down come from its timeline instead of memory.
Start monitoring free →Related terms
Part of the Uptimely monitoring glossary — plain definitions of the vocabulary uptime monitoring, incident response and status pages are discussed in. Something here read wrong to you? Email support@getuptimely.com.