Prometheus Operators in the Wild
summary
Compares two OSS operators and shows a live demo on a 3‑node cluster with three collectors; one operator runs as a DaemonSet per node (GMP) and limits scrape scope to each node, the other uses hashmode shards to split targets across a chosen number of Prometheus instances. Both use podMonitor / PodMonitoring CRDs for per‑app scrape config and send metrics via remote_write to Prometheus receiver. Demo deploys both, they scrape an app and successfully forward metrics; GMP is simpler and GKE‑enabled, Prometheus Operator is more feature‑rich.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 1:27Speakers
- 2:54Why use Prometheus operators
- 5:44Benefits in shared envs
- 7:33Simpler scrape config
- 10:00Prometheus Operator overview
- 12:06GMP Operator intro
- 13:30Key GMP CRDs
- 15:18Optimize GMP as DaemonSet
- 16:45Live demo setup
- 18:10Deploy Prometheus Operator
- 18:31Deploy GMP Operator
- 19:36Configure scrape targets
- 20:20Verify metrics
- 21:22Compare operators
- 24:51Choosing operators
- 25:35Q&A
related talks
31mDeep Dive into the Prometheus Operator Ecosystem
24mHow to Be 10x SRE? A Deep Dive to Prometheus Operator
26mEnd-to-end monitoring with the Prometheus Operator
35mDeclarative Multi-Cluster Monitoring with Prometheus
35mStateless Collectors For Stateful Data: Scaling Prometheus As a Node Agent
02h 43mPrometheus from A to Z
34mMonitoring Spinnaker with Prometheus Operator on GKE
29mHarnessing the Potential of Prometheus Agent Mode