Ignite | Discovering the Magic Behind OpenTelemetry Instrumentation
summary
Claims that manual instrumentation beats auto-instrumentation for correctness and cost control, especially in languages like C++ where monkey‑patching isn’t possible. Uses OpenTelemetry API to emit metrics, spans and logs to an SDK that aggregates and exports in batches (seconds), with exporters for Prometheus or other backends. Context propagation follows the W3C spec and reusable semantic conventions standardize attributes. Notes that the collector can synthesize collector-generated metrics from spans and that high-cardinality data quickly raises performance and cost.
AI-written summary · about
related talks
31mDiscovering the Magic Behind OpenTelemetry Instrumentation
23mModern Application Debugging: An Intro to OpenTelemetry
22mFine-Tuning Auto-Instrumentation
43mWhat is this OpenTelemetry thing?
20mObservability and Open Telemetry
36mObservability For You and Me with OpenTelemetry
46mPanel: Meet The Maintainers
01h 28mUnderstanding Distributed Systems with OpenTelemetry