Efficient Python Observability with OpenTelemetry
summary
Argues for a minimal, practical observability setup using OpenTelemetry for Python: uses auto-instrumentation wrapper for quick coverage, framework integration (Flask) for low-code tracing, and manual spans via with blocks for precise hotspots. Demonstrates batching with a batch span processor to reduce network calls and shows filtering spans to export only relevant traces. Exports telemetry to Jaeger/Zipkin, claims measurable reduced noise and acceptable overhead while making distributed request paths and DB calls visible.
AI-written summary · about
chapters (6)
- 0:00Introduction and telemetry basics
- 3:52Why OpenTelemetry matters
- 7:22OpenTelemetry architecture
- 10:33Python examples and demos
- 15:48Viewing telemetry results
- 17:13Q&A
related talks
29mModern observability for Python
27mDistributed Tracing with OpenTelemetry
24m"Introduction to OpenTelemetry with Python"
38mEnhance your Python application observability
24mObservability and instrumentation via OpenTelemetry
23mGetting Started with OpenTelemetry in Python
30mObserving Python applications with OpenTelemetry
30mPractical introduction to OpenTelemetry Tracing