Skip to content

πŸ“ˆ Service Level Indicator (SLI) – Deep Dive

What is an SLI?

A Service Level Indicator (SLI) is a carefully defined quantitative measure of some aspect of the level of service that is provided. SLIs are the foundation of SLOs and SLAs, and they help in assessing whether users are experiencing the desired reliability and performance.

SLI is a specific measurement that tells you how well a service is working from the user’s point of view.

Characteristics of a Good SLI

  • User-Centric: Reflects the user's experience with the service
  • Quantifiable: Can be measured with real metrics
  • Relevant: Maps to critical service behavior
  • Actionable: Enables alerting, improvements, or incident response

Common SLI Types

  • Availability – How often is the service reachable?
  • Latency – How fast is the service?
  • Error Rate – How frequently does it fail?
  • Durability – Is data safe and not corrupted?
  • Freshness – Is data up-to-date?

SLI Formula Example

SLI = (Number of successful requests / Total requests) * 100

πŸ—‚οΈ Example SLIs by Service Type

Service Type Example SLIs
Web APIs / Backend Services HTTP success rate, request latency, error rate
User-Facing Web Applications Page load time, JS errors/session, navigation success rate
Mobile Apps API latency, crash rate, sync latency
Authentication Services Login success rate, authentication time, timeout rate
Data Pipelines / ETL Jobs Data freshness, pipeline success rate, processing delay
CI/CD Services Pipeline provision success, execution time, rollback rate
Messaging / Notification Systems Message delivery rate, latency, duplication/drop rate
Database Services Query success rate, latency, transaction failure rate
File/Storage Services Retrieval success rate, download latency, data corruption rate
Streaming / Real-time Services Stream start time, buffering ratio, dropped frames
Search Services Search latency, result relevance, query failure rate
Email / Communication Services Delivery time, bounce rate, spam/marked message rate