OpenTelemetry's Path to Observability's Future
summary
Claims around 4–6 tools per company for telemetry collection and that OpenTelemetry is the CNCF project ranked second most active provide context. The architecture centers on the open telemetry collector (receivers → processors → exporters) with examples exporting to Zipkin, Prometheus and New Relic OTP. Concrete outcomes: tail-based sampling that only keeps traces over 5s meaningfully cuts ingestion and cost, and a SkyScanner case that retires 12 systems and saves ~15 minutes per mobile build step.
AI-written summary · about
chapters (17)
- 0:00intro
- 1:43preamble
- 1:58whoami
- 2:17agenda
- 2:33how many tools companies use to collect telemetry data?
- 2:51monitoring vs observability
- 4:03why observability
- 5:10opentelemetry
- 7:11second most active cncf project
- 7:26core concepts
- 8:36instumentation
- 9:40collector
- 14:06sampling config
- 15:30way to export
- 16:18industry adoption
- 18:27recap and highlights
- 19:11reach out, thanks!
related talks
20mObservability and Open Telemetry
18mOpenTelemetry and Observability: What, Why, and Why Now?
09mOpenTelemetry or OpenTeleMessy?
15mOpenTelemetry: The Vision, the Reality and How To Get Started
34mOpenTelemetry: The Vision, Reality, and How to Get Started
22mOpenTelemetry – An Observability Framework for Cloud-Native Software
47mWhat Is This OpenTelemetry Thing?
01h 08mOpenTelemetry: The First Release, What’s Next, and How to Get Involved