ServiceLevelObjectives

Service Level Objectives (SLOs) are formal targets for the reliability and performance of a service, stated in terms of a measurable service level indicator (SLI) over a defined time period. An SLO specifies the level that the service aims to maintain, such as system availability or response time, and is used to assess whether the service meets its reliability goals. The SLO concept originated in site reliability engineering (SRE) and is widely used to translate user expectations into concrete targets. The SLI is the metric used to gauge performance; the SLO is the desired target for that metric.

Examples of common SLOs include uptime of 99.9% per calendar month, p95 latency under 200 milliseconds, and error rate below 0.1%. SLOs should be specific, measurable, attainable, relevant, and time-bound. They are typically monitored with dashboards and alerting tied to the underlying SLIs.

SLOs are distinct from service level agreements (SLAs). SLAs are external commitments, often accompanied by penalties or remedies if targets are not met, whereas SLOs are internal goals used to guide development and operations. An associated concept is the error budget, the amount of allowable deviation from the SLO within a period, which helps balance reliability with feature development.

Operations use SLOs to inform incident response, capacity planning, prioritization, and continuous improvement. They influence alert thresholds, post-incident reviews, and resource allocation, and are most effective when aligned with user expectations and business objectives.