Is It Already Time To Version Observability? (Signs Point To Yes.)
summary
Argues observability needs a major-version bump: Observability 2.0 replaces fragmented metrics+logs+traces with a single source of truth of wide structured events, stored raw indefinitely in a columnar engine (eg. ClickHouse), enabling high cardinality, read-time aggregation and exploratory UIs. Claims concrete outcomes: you can delete 90–95% of paging alerts by aligning alerts to SLOs, trace with span IDs in the same event stream, and pay to store once rather than incur exploding per-metric bills. Recommends canonical logs, columnar storage, and explorable interfaces.
AI-written summary · about
chapters (16)
- 0:00Introduction
- 1:06Defining observability
- 3:35Observability as system property
- 5:20Observability 1.0 vs 2.0
- 9:16Data structures and tooling shifts
- 11:00Observability 2.0 impact
- 14:34Observability beyond debugging
- 17:40Debugging workflows 1.0 vs 2.0
- 20:11Cost crisis in observability
- 24:05Metrics logs and traces
- 27:17Why metrics are insufficient
- 29:05Adopting observability 2.0
- 31:54Storage engines and cardinality
- 33:40Complexity driving observability 2.0
- 35:48Future of engineering and observability
- 38:17Conclusion
related talks
35mIs It Time To Version Observability? (Signs Point To Yes)
33mKeynote: Observability is due for a version change; Are you ready for it?
52mObservability: it's not just an ops thing
31mObservability for Developers: How to Get There
39mThe 5W’s of Observability with Practical Examples
29mObservability; a gentle introduction
36mObservability? It's a Data Analytics Problem, You Fool!
34mThe State of SQL-Based Observability