From Blind Spots to Total Vision: O11y at Massive Scale
summary
Argues that lowering mean time to detect is the key goal and shows a working pipeline: OpenTelemetry agents and Fluent Bit collect traces and logs, metrics go to Prometheus/Grafana, OpenSearch stores logs/traces and scales to ~25 PB. Demo generates failures in a payment service that produce 11 error traces, uses trace ID filtering to pull correlated logs and finds a 503 root cause. Shows ML features: anomaly grade and confidence (suggested cutoff 0.5) and a zero‑ETL CloudWatch→OpenSearch integration.
AI-written summary · about
chapters (9)
- 0:00Introduction and Speaker Background
- 0:22Understanding Observability
- 2:01Components of Observability
- 3:40Introduction to OpenSearch
- 4:32OpenSearch Use Cases and Deployment
- 7:00Instrumentation and Data Collection
- 10:30Analyzing Data with OpenSearch
- 17:37Demo: Using OpenSearch for Observability
- 22:07Q&A Session
related talks
20mSeeing Is Believing: OpenTelemetry and OpenSearch in Next-Gen Observability
11mObservability 2.0 - Much more than just logs,metrics and traces
20mSuper vision smart goggles
21mFrom Black Box to Glass Box: AI-Driven Observability for Modern APIs
45mFrom Noise To Signal: Building Smart Observability Pipelines With OpenTelemetry
27mAlert Fatigue to Action: Observability as the Backbone of Practical AIOps with OpenSearch
26mYou Can’t Fix What You Can’t See: Tales from the Observability Frontier
21mDebugging at Scale—Going from Single Box to Production