Skip to content

Level 4 · SLOs & Operating Observability at Scale

Level 4 steps back from any one tool to the operational practice that ties them all together: turning "is the system healthy?" into a number you can actually set a target for, alert on sensibly, and use to decide what to fix first. This level also covers what breaks about this course's single-node setup at real scale, honestly.

What was run, and what wasn't

This level computes real SLO math and real burn-rate alert thresholds against real metrics from the services built earlier in this course, and runs a real capacity-planning projection from real measured data. Scaling Prometheus itself to a multi-region, long-term-storage deployment (Thanos, Cortex, Mimir at real production scale) is covered as documented architecture, not run locally, and that's stated plainly.

Prerequisites: Levels 1–3.

Modules

  1. SLIs, SLOs & SLAs: the Real Differences: three terms used interchangeably that mean genuinely different things, with a real SLI computed from real data.
  2. Error Budgets: Turning Reliability Into a Spendable Number: a real error budget, computed and tracked against a real SLO over time.
  3. Burn Rate Alerts: Multi-Window Multi-Burn-Rate Alerting: a real threshold alert compared against a real burn-rate alert on the same real incident.
  4. Designing an SLO for a Real Service: a full, real SLO designed from scratch for a service built earlier in this course.
  5. Capacity Planning From Real Metrics: a real growth projection computed from real historical metric data.
  6. Scaling Prometheus: Federation, Remote Write & Long-Term Storage: what actually breaks past one node, and the documented architectures that fix it.
  7. Observability as Code: Dashboards & Alerts in Version Control: a real dashboard and alert rule, defined as code and diffed like any other change.
  8. On-Call & Incident Response With Your Own Stack: a real injected incident, worked end to end using only this course's own stack.
  9. Common Observability Anti-Patterns: real, measured costs of getting this wrong, tallied from across the whole course.
  10. Capstone: Build a Complete SLO-Driven Observability Stack: metrics, dashboards, alerts, traces, and an SLO, for one real service, end to end.

After Level 4

You've instrumented a real service, built real dashboards and alerts, traced a real request across process boundaries, and defined a real SLO to operate it by — the same loop a production observability practice runs continuously, just at a scale you can finish on a laptop.