SLO/SLI Design Framework for Distributed Systems

You are an SRE Lead designing reliability contracts for high-traffic microservices.

Context

We need to define the reliability standard for the following journey: {{user_journey}}. The business requirement is {{availability_target}} uptime with a latency target of {{latency_threshold}}.

Task

1. Identify the specific SLIs (Service Level Indicators) that best represent the user experience for {{user_journey}}.
2. Define the 'Good Events' vs. 'Total Events' for each SLI.
3. Calculate the monthly Error Budget in seconds/minutes based on {{availability_target}}.
4. Propose alerting thresholds (Burn Rate alerts) for 1-hour and 6-hour windows.
5. Select the appropriate metrics from Prometheus/Datadog (e.g., http_requests_total) to track these SLIs.

Constraints

• MUST focus on user-centric metrics, not just infrastructure health (CPU/RAM).
• MUST provide the mathematical formula for the SLI.
• MUST specify the 'window' for the SLO (e.g., 28-day rolling window).

Output format

• Service Level Indicator (SLI) Definitions
• SLO Specification Table (Indicator, Objective, Measurement Window)
• Error Budget Analysis
• Burn Rate Alerting Logic (Table: Alert Level, Window, Threshold)

Quality bar

• Are the SLIs measurable and actionable?
• Does the error budget calculation mathematically align with {{availability_target}}?
• Are the burn rate alerts designed to minimize alert fatigue?