MTTR — Mean Time to Recovery / Repair / Resolve

• MTTR measures the average time required to restore or resolve a service after a failure.
• Lower MTTR generally indicates faster incident recovery.
• Depending on the organization, MTTR may mean:
  • Mean Time to Recovery
  • Mean Time to Repair
  • Mean Time to Resolve

MTBF — Mean Time Between Failures

• MTBF measures the average operating time between failures of a system.
• Higher MTBF generally indicates fewer failures.
• It is commonly used as a measure of system reliability.

MTTD — Mean Time to Detect

• MTTD measures the average time required to detect that an incident or failure has occurred.
• Lower MTTD means incidents are detected more quickly.
• Monitoring, alerting, and observability help reduce MTTD.

MTTA — Mean Time to Acknowledge

• MTTA measures the average time between an alert or incident being reported and someone acknowledging it.
• Lower MTTA means teams acknowledge incidents more quickly.
• It is commonly associated with incident response.

SLA — Service Level Agreement

• SLA is a formal agreement between a service provider and customer that defines expected service levels and commitments.
• An SLA may specify:
  • Availability
  • Response time
  • Support commitments
  • Resolution expectations
• Example: A provider may commit to 99.9% monthly availability.

SLO — Service Level Objective

• SLO defines a specific target for a service's reliability or performance.
• SLOs are typically used internally to guide engineering and operations.
• Example: 99.9% availability over a given period.

SLI — Service Level Indicator

• SLI is the actual measurement used to evaluate a service's performance or reliability.
• Common SLIs include:
  • Availability
  • Latency
  • Error rate
  • Throughput

SLI vs SLO vs SLA

• SLI → What do we measure?
• SLO → What target do we want?
• SLA → What do we promise the customer?

Example

• SLI: Actual availability = 99.95%
• SLO: Target availability = 99.9%
• SLA: Customer commitment = 99.9%

Incident Management Metrics

• MTTD → How quickly did we detect the incident?
• MTTA → How quickly did we acknowledge the incident?
• MTTR → How quickly did we recover or resolve the incident?
• MTBF → How long did the system operate before another failure?