Service Level Terminology: SLI, SLO, & SLA

The Three Concepts Defined

 ┌────────────────────────────────────────────────────────┐
 │ SLI (Service Level Indicator)                          │
 │ "A carefully defined quantitative measure of service"   │
 │ Example: Successful HTTP requests / Total HTTP requests │
 └───────────────────────────┬────────────────────────────┘
                             │
                             ▼ Target Threshold
 ┌────────────────────────────────────────────────────────┐
 │ SLO (Service Level Objective)                          │
 │ "A target value or range for an SLI agreed internally"  │
 │ Example: 99.9% of requests must succeed over 30 days   │
 └───────────────────────────┬────────────────────────────┘
                             │
                             ▼ Business / Contractual
 ┌────────────────────────────────────────────────────────┐
 │ SLA (Service Level Agreement)                          │
 │ "An explicit contract with users outlining consequences│
 │ Example: If availability drops below 99.5%, refund 10% │
 └────────────────────────────────────────────────────────┘

1. SLI (Service Level Indicator)

A factual, real-time metric measured over a window of time: $$\text{SLI} = \frac{\text{Good Events}}{\text{Total Events}} \times 100%$$

Request-driven services: Availability (HTTP 2xx vs 5xx), Latency (95th/99th percentile < 200ms).
Data-processing pipelines: Freshness, correctness, throughput.
Storage: Durability, availability, latency.

2. SLO (Service Level Objective)

The target reliability level set by product management and SRE teams jointly:

E.g., “99.9% of /checkout HTTP requests respond in < 300ms over a rolling 30-day window.”
Crucial rule: SLO should always be tighter than SLA ($\text{SLO} > \text{SLA}$) to provide an operational buffer before financial penalties occur.

3. SLA (Service Level Agreement)

The external business or legal contract signed with paying customers:

Contains financial penalties, subscription credits, or termination clauses if breach occurs.
SREs rarely interact with SLAs directly; their daily work centers on maintaining SLOs and managing Error Budgets in SRE.