Incremental Strategies for Adopting OpenTelemetry
summary
Describes an incremental OpenTelemetry adoption plan: ingest existing logs and metrics using the available ~90 receivers, enable auto/zero‑code instrumentation for fast coverage, and add explicit SDK spans where they provide value. Recommends a collector-per-node plus gateway hierarchy (DaemonSet or systemd on VMs), use batching and local file or Kafka buffering to prevent loss, monitor collector metrics via an exporter, and build a slim collector image with OpenTelemetry Collector Builder instead of heavy third‑party distributions; notes OAMP is still nascent.
AI-written summary · about
chapters (5)
- 0:00OpenTelemetry intro
- 4:41Telemetry data collection
- 20:40Incremental adoption
- 33:07Best practices
- 35:14Q&A
related talks
37mHow To Adopt OpenTelemetry in an Enterprise Where Incumbent Vendor Tools Reign Supre...
36mObservability For You and Me with OpenTelemetry
35mTales of Adopting OpenTelemetry: A First-Timer Journey
15mOpenTelemetry: The Vision, the Reality and How To Get Started
30mOpenTelemetry 101
32mOpenTelemetry feat. logs, metrics and traces
26mScaling OpenTelemetry for Modern Workloads: From Mobile To Ma...
52mHow to effectively spy on your systems