Definition: Quantitative metric measuring a specific aspect of service level, such as latency or error rate, serving as the foundation for defining SLOs and SLAs.
— Source: NERVICO, Product Development Consultancy
What Is an SLI
A Service Level Indicator (SLI) is a carefully chosen quantitative metric that measures a specific aspect of the service from the user’s perspective. It is the raw data upon which SLOs and ultimately SLAs are built. Typical SLIs include response latency, request success rate, throughput, and service availability.
How It Works
An SLI is defined as a proportion of good events over total valid events. For example, the latency SLI could be defined as “the percentage of HTTP requests completed in under 300ms.” The availability SLI would be “the percentage of requests returning a successful status code (2xx or 3xx).” These indicators are continuously collected using observability tools like Prometheus, CloudWatch, or Datadog, and aggregated over defined time windows to evaluate SLO compliance.
Key Use Cases
- Continuous measurement of actual user experience through latency and error rate indicators
- Quantitative basis for defining realistic SLOs aligned with user expectations
- Powering reliability dashboards showing error budget status in real time
- Automatic detection of service degradation through SLI threshold-based alerts
Advantages and Considerations
SLIs provide an objective and quantifiable measure of service quality, eliminating subjective evaluations. By being defined from the user’s perspective, they ensure metrics reflect actual experience. The main consideration is selecting the right SLIs: too many indicators disperse attention, while poorly chosen ones can obscure real problems. Starting with few high-impact SLIs and gradually expanding is recommended.