Enabling Intelligent Observability Volume Management
Talkat Observability Day North America 2024 · on Tuesday, 2024-11-12· at 16:30· Salt Palace | Level 2 | 255 E·
summary
OVM is an open source pipeline component that sits between collectors and central aggregators and performs dynamic, edge‑side transformations such as filtering, aggregation and per‑metric scrape‑frequency changes. It integrates with Prometheus and OTel, runs a processor at each edge plus a central manager/controller that recommends and auto‑applies rules, demonstrably drops redundant metrics (demo: filter by IP) and reduces anomaly‑function impact by ~600x while keeping dashboards and analytics intact.
AI-written summary · about
chapters (5)
- 0:00Intelligent observability intro
- 5:59Traditional observability challenges
- 10:54OVM solves volume issues
- 16:51OVM benefits for stacks
- 20:20OVM future plans
related talks
11mObservability 2.0 - Much more than just logs,metrics and traces
11mLightning Talk: OTel Me How to Build a Data Pipeline for Observability
37mObservability For You And Me With Opentelemetry
48mAPI Observability by Design: Building Systems That Tell Their Own Stories
22mAI-Driven Observability and Operations in Cloud-Edge Systems
23mObservability in OpenStack Learnings and building blocks from the SIG Monitoring at SCS
31mObservability: Dust Off Prometheus with VictoriaMetrics
53mOpenTelemetry Metrics: observability’s holy grail