One Pane to Rule Them All: Uniting the Prometheus Community with O...
summary
OpenSearch now ingests Prometheus metrics natively and surfaces them alongside logs and traces in OpenSearch Dashboards using Data Prepper and OpenTelemetry collectors, so multiple Prometheus clusters query with PromQL and combine with PPL without changing engines or alert managers. The demo shows unified SLOs, alerts, application maps, and trace-log correlation working on live clusters; the project has ~4,000 community members, a new ECharts viz stack (consolidating ~15 libs in six months), and Prometheus as first-class citizen and no data migration, no engine change approaches.
AI-written summary · about
chapters (11)
- 0:00Prometheus and OpenSearch integration
- 4:30Prometheus metrics in OpenSearch
- 7:37Seamless integration workflows
- 9:02Demo: metrics in OpenSearch
- 12:29Unified alerts and SLOs
- 15:19Logs, traces, metrics dashboards
- 17:26Future of observability with AI
- 18:29Getting started with OpenSearch
- 20:23Q&A: scaling OpenSearch clusters
- 22:21OpenSearch visualization improvements
- 23:48Closing and next steps
related talks
25mUnifying Observability: Weaving Prometheus, Jaeger, and Open Source Together to Win
11mOpenTelemetry + OpenSearch Integration
10mPrometheus for Monitoring Metrics - Database Monitoring Tools
29mParanoid Observability –Success Stories and Misfortunes observing an observability stack
37mBuilding Modern Observability Architecture With OpenSearch, OS Ingest...
20mExtending OpenSearch Dashboards: Building a Unified Obser...
13mConnecting Prometheus and OpenTelemetry Data for Faster Troublesho...
27mObservability with Prometheus and beyond