SRE Core Triad: SLIs, SLOs, and SLAs

The foundation of Site Reliability Engineering relies heavily on data-driven management. Rather than guessing whether a system is reliable, SRE targets objective mathematical definitions.

The Triad at a Glance:

SLI (Service Level Indicator): The quantitative measure of performance (e.g., Error Rate).
SLO (Service Level Objective): The target metric specified for reliability (e.g., Error Rate < 0.1%).
SLA (Service Level Agreement): The legal commitments made to users, including financial consequences if breached (e.g., If SLO is missed, clients receive a 15% refund).

Visualizing the Reliability Relationship

The chart below displays how operational indicators feed internal organizational goals, which ultimately shield binding customer agreements:

graph TD A[SLI: Real-time Metric Production] -->|Aggregated & Monitored| B(SLO: Internal Reliability Target) B -->|Safeguards Against Breaches| C(SLA: External Legal Agreement) style A fill:#1e293b,stroke:#38bdf8,stroke-width:2px,color:#fff style B fill:#1e293b,stroke:#a855f7,stroke-width:2px,color:#fff style C fill:#1e293b,stroke:#ef4444,stroke-width:2px,color:#fff

1. Service Level Indicators (SLIs)

An SLI is a foundational building block. SRE structures indicators as a ratio comparing successful events against total valid events:

                        SLI = ( Good Events / Total Events ) × 100
                    

For example, a common availability indicator tracks HTTP responses: the number of HTTP 200 status codes divided by all returned routing signals.

2. Service Level Objectives (SLOs)

The Service Level Objective represents the target boundary your infrastructure is expected to preserve over a specific rolling timeline window (e.g., 30 days). Maintaining an SLO involves defining exactly what failure looks like.

SRE methodology actively discourages aiming for 100% reliability because users cannot distinguish between 99.9% and 100% availability. For example, localized network connections and background carrier interference cause small drops in connectivity regardless of data center stability.

3. Service Level Agreements (SLAs)

While SLIs and SLOs are technical metrics, the Service Level Agreement is a legal and commercial contract. It describes the consequences of missing the target reliability. If a company fails to meet its SLA, it typically faces financial penalties, service credits, or contract termination.

SRE Tip: Keep your SLO targets stricter than your SLA parameters. This gives your engineering team an early warning system to fix reliability issues before they trigger costly contractual violations.