What is a recovery time objective (RTO)?

DefinitionRecovery time objective (RTO)

The recovery time objective is the maximum time a system may stay unavailable after an incident before the disruption becomes unacceptable. It sets how quickly a service has to be restored and running again.

At PilaePilae probes every app every 60 seconds, and alerts reach engineers around the clock.

TopicBackups and recovery

Written by Paul Madelénat.

Talk to an engineerBackups and recovery

On this page (3)

What the recovery time objective measures

The recovery time objective is a deadline. The clock starts when a system stops serving its users and stops when they can work with it again. An RTO of four hours means the organisation has decided that four hours without the system is tolerable and five is not.

RTO covers the whole recovery, not only the restore itself. That is why detection and a clear runbook matter as much as fast storage.

What the RTO clock covers

  1. Notice: the failure is noticed.
  2. Decide: what to do.
  3. Prepare: a machine is made ready.
  4. Restore: the data is restored.
  5. Check: the app answers.

It is often confused with two neighbouring ideas. The recovery point objective measures how much data may be lost, not how long the outage lasts. An availability commitment, such as a monthly percentage, measures uptime over a period. A single long outage can respect the monthly figure and still break the RTO, so the two are set separately.

How to set an RTO

Ask what happens to the people who depend on the system, hour by hour, when it is gone. Email and file sharing usually need a short RTO. An internal reporting tool can often wait until the next day. The shorter the RTO, the more it costs: standby machines, replicated data and people ready to act at any hour.

Then test it. Time a real restore of each app into a clean environment and compare the result with the objective. If the restore takes longer than the RTO, either the objective or the setup has to change.

RTO at Pilae

Pilae probes every app every 60 seconds, and alerts reach engineers around the clock. Work starts within the response target of your plan, and around the clock for Severity 1 and 2 incidents on Enterprise. Each restore follows a recorded procedure: an engineer confirms the backup and the expected downtime, restores the app, checks it answers and writes the duration and result to the audit log.

On Enterprise, your contract states the RTO for each app, alongside the monthly availability commitment and its service credits, and monthly restore tests, recorded in the console, show whether the objective holds. On Business, the availability commitment and its credits apply without a per-app objective. See backups and recovery and monitoring, or agree your objectives with an engineer.

Talk to an engineer about your backups and restore tests

Thirty minutes with an engineer, a written plan and a fixed price for the first workload.

Book 30 minutes

  • Go deeperBackups and recoveryDaily encrypted offsite backups on every plan, longer retention and point-in-time recovery for databases from Business, monthly restore tests and recovery objectives on Enterprise.
  • CompareWhat is a recovery point objective (RPO)?The recovery point objective is the maximum amount of data, measured in time, that an organisation accepts to lose after an incident.
  • CheckMonitoringProbes every 60 seconds, machine metrics and alerts that reach Pilae engineers around the clock, with a monthly report.

Tell us what you need to run.