SRE Jumpstart
summary
Covers concrete Site Reliability Engineering practices: defines customer‑user journeys (CUJs), selects specific SLIs/SLOs and shows the math that a 99.9% SLO gives a 0.1% error budget (~7.2 hours/month) translated into minutes and operational decisions. Demonstrates a Terraform-based demo shop and load generator on Google Cloud to measure SLIs, explains risk‑ranking by time‑to‑fail and time‑to‑repair to compute bad‑minutes and prioritize fixes, and describes enforcement mechanisms — dashboards, alerts, and error‑budget driven controls such as feature freezes and code‑red procedures with real examples of triggers.
AI-written summary · about
chapters (24)
- 0:00Welcome
- 1:03SRE and DevOps intro
- 3:09SRE origins and principles
- 6:17Google's three SRE principles
- 7:42DevOps vs SRE differences
- 9:48Opinionated DevOps
- 11:56SLO definition and importance
- 13:20Calculating SLO
- 15:26Google's SLO calculation
- 17:34Practical SLO calculation
- 18:38Live demo: service analysis
- 20:20SLO in context
- 22:46Error budget definition
- 25:17Using error budget
- 27:45Documenting SLO
- 29:52Monitoring dashboards
- 31:56Emergency handling
- 33:39Risk analysis and improvement
- 36:30Error ranking
- 38:18SLO updates
- 39:29SRE tips
- 40:12Q&A 1
- 43:06Q&A 2
- 45:34Conclusion
