Keynote: Prometheus 2.0 – The Next Scale of Cloud Native Monitoring

on YouTube ↗16m

" summary

Prometheus 2.0 replaces the TSDB with a new storage engine and open library that compresses series, lowers baseline memory/CPU, and makes ingestion far cheaper; benchmarks show 1 million samples/sec per CPU core, up to two orders of magnitude ingestion improvement and storage savings up to 4x in dense, short-lived-series workloads. Pull-based termination markers let Prometheus mark series as ended so alerts and graphs reflect termination within one scrape interval instead of minutes.

AI-written summary · about

" chapters (7)

1. 0:00Start
2. 6:40Integrating TSDB
3. 7:03Optimizing Scraping
4. 7:25Benchmarks
5. 11:13Rule Format
6. 11:41Staleness
7. 12:50What's next?

" related talks

• 32mPrometheus Deep DiveBen Kochie · KubeCon + CloudNativeCon North America 2019
• 13mLightning Talk: A Drop-in System To Accelerate Metrics Observability by 100x Us...Milind Srivastava · Observability Day Europe 2026
• 35mStateless Collectors For Stateful Data: Scaling Prometheus As a Node AgentDanny Clark · KubeCon + CloudNativeCon North America 2022
• 22mThat's Not a Lot of Data; Leveraging Cloud-native Technologies to Collect, Stor...Richard Hartmann · Open Source Strategy Forum London 2021
• 04mLightning Talk: Prometheus Native Histograms Are Now a Stable FeatureKrajo Krajcsovits · PromCon EU 2025
• 35mSustainability Research the Cloud Native WayChen Wang, Huamin Chen · KubeCon + CloudNativeCon North America 2022
• 24mOne Pane to Rule Them All: Uniting the Prometheus Community with O...Anirudha Jadhav, Kevin Fallis · Observability Summit North America 2026
• 12mKeynote: Cloud Native Sustainability - How You Can Help Right Now!Saiyam Pathak, Nancy Chauhan · KubeDay India 2023