What Developers Should Know About Observability
summary
Covers practical claims: four pillars (metrics, logs, traces, profiling) and why profiling gives line‑level hotspots while metrics give high‑level health. Describes tools and tradeoffs: Prometheus for metrics, OpenTelemetry for tracing but requiring code changes, and eBPF for agentless, dynamic probes that map service‑to‑service without instrumentation. Demonstrates a product that uses ClickHouse for cost‑efficient storage, detects cloud billing and cross‑AZ waste, and produces AI suggestions that are useful about 80% of the time.
AI-written summary · about
chapters (8)
- 0:00Observability importance
- 3:56Why observability matters
- 8:33Four pillars explained
- 12:34Instrumentation and data collection
- 17:35Service level objectives
- 20:06Common observability challenges
- 24:46Karut’s observability approach
- 31:11Effective observability advice
related talks
34mObservability Concepts Every Developer Should Know
31mObservability for Developers: How to Get There
35mBeyond Logs, Metrics and Traces. The Data Analytics Paradigm Shift in Observability — Dotan Horovits
48mAPI Observability by Design: Building Systems That Tell Their Own Stories
16mBuilding Observability into Cloud-Native Applications
29mObservability; a gentle introduction
44mObservability: Superpowers for Developers
32mUnderstanding Observability in Kubernetes