Monitoring OVH: 300k Servers, 27 DCs, a Metrics Platform
summary
OVH Metrics centralizes monitoring across 27 datacenters and over 300,000 physical servers, ingesting millions of time series per minute and serving 1.3 million customers. The platform uses distributed collectors, push proxies, and long‑term storage on capacity‑stressed HBase clusters, migrates legacy silos and data from Graphite and Prometheus frontends, and delivers multi‑tenant metrics and large‑scale alerting in production.
AI-written summary · about
chapters (9)
- 0:00OVH monitoring intro
- 3:57300k servers 27 DCs
- 8:35Metrics platform evolution
- 15:11Monolith to microservices
- 20:42Prometheus Grafana today
- 26:05Scaling monitoring solutions
- 31:10Tool adoption challenges
- 36:16Case studies lessons
- 40:28Conclusion future
related talks
32mMonitoring OVH: 300K servers, 27 DCs one unified metrics platform
36m200 Clusters vs 1 Admin
22mImproving the efficiency of more than 800 databases with o11y
17mSRE implementation in a cool company
43mMonitoring at one of the largest retail groups in the world
45mDecentralizing SaaS Monitoring Using Open-Source Tools and Standards
36mAdventures in Running a Private Cloud
25mHelp, My Datacenter is on Fire