Observability Diet: Your 5-Step Plan To Trim the Data Fat
Lightningat KubeCon + CloudNativeCon Europe 2025 · on Thursday, 2025-04-03· at 16:55· Level 0 | ICC Auditorium·
summary
Many orgs send terabytes of telemetry while using only 20% of it, so the plan focuses on concrete volume cuts: run OpenTelemetry Collector processors to tail-based sampling and head‑based or probabilistic sampling for logs and traces, drop irrelevant attributes (for example drop http.headers at the SDK or collector), use SDK Views in SDKs to reduce metric cardinality, filter logs by severity, and apply granular retention to slash costs while keeping the signals that actually drive alerts and dashboards.
AI-written summary · about
related talks
16mShrinking the Observability Bill
22mShift Into an Observability Mindset with OpenTelemetry
30mPanel: Telemetry That Matters
02h 41mThe Science of Signals: Mastering Telemetry for Observability
30mBut Wait! There's...Still More‽ - Observability Data Volumes and Strategies for Managing Them
23mControlling Data Overhead with the OpenTelemetry Collector
35mBeyond Logs, Metrics and Traces. The Data Analytics Paradigm Shift in Observability — Dotan Horovits
35mRefining Systems Data without Losing Fidelity