Lightning Talk: Cognitive and SelfAdaptive System for Distributed Tracing
summary
Claims a machine‑learned, tail‑based sampler layered over Jaeger/OpenTracing that replaces random 1–5% head sampling with a density‑aware semi‑supervised model (DBSCAN style) that builds feature vectors (service, path, status, duration, host), clusters traces via Kafka ingestion, assigns inverse weights, and retains rare/high‑interest traces. Storage drops from petabyte‑scale growth when raising rates, adaptive rates cut noise and cost, and reported results show a significant MTTR reduction for SREs. tail‑based sampling adaptive sampler
AI-written summary · about
chapters (6)
- 0:00Distributed tracing intro
- 1:51Tracing components basics
- 3:39Traditional sampling problems
- 5:45Tail-based adaptive sampling
- 9:01Adaptive sampler engine
- 10:53Q&A
related talks
14mCognitive and SelfAdaptive System for Distributed Tracing
13mDistributed Tracing: Adaptive and Telemetry-Based Approach for Effective...
11mCognitive and Self-Adaptive System for Effective Distributed-Tracing...
13mCognitive and Self-Adaptive System for Effective Distributed-Tracing
21mLightning Talk: Distributed Tracing Without the Pain!
31mDistributed Application Tracing with Trace4Cats
12mLightning Talk: With Great Traces Comes Great Costs: How to Reduce That Bill?
45mMastering Distributed Tracing: Insights from Building a Platform from Scratch