The Frontiers of Reliability Engineering
summary
Zalando runs ~3,500 engineers and more microservices than people, deploys on AWS + Kubernetes, uses Docker, Prometheus/Grafana and OpenTelemetry, and claims tracing investment of ~5 years yields 250 spans per request (sometimes up to 1,000) that drive SLOs and alerting. Management uses a Python-generated reliability report (Google Doc) and weekly operational reviews plus an on‑call health table with a red threshold at 10 pages. Mobile releases hit 80% in ~4 weeks; a data bug forced a 3.5‑hour full reindex.
AI-written summary · about
chapters (7)
- 0:00Introduction and context
- 2:33A decade of SRE evolution
- 8:02The impact of observability
- 9:56Core reliability principles
- 16:27Managing for reliability
- 23:40Mobile observability challenges
- 30:24Data operations and future outlook
related talks
14mWeb Apps, through the Lens of SRE (Site Reliability Engineering)
41mBuilding and running applications at scale in
43mThe Invisible Door: Reliability Gaps in the Front End
13mBeyond observability: Using reliability scores to drive results
40mThe Recipe For Scalable Frontends
32mBeyond Optimization: Resilient Cloud Microservices at Scale
35mThe elements of scaling
54mFrom Code to Customer