Deep Dive: Prometheus
summary
Prometheus explains concrete engineering: a new TSDB reduces resource use dramatically in benchmarks with 800 microservices, ~100–120k scrapes/sec and ~300k active series, showing 15x memory reduction, 6x CPU drop and up to 100x fewer disk writes. It stabilizes remote read/write for long-term export, adds RC-driven six-week releases, improves staleness/latency handling, and plans TLS for exporter endpoints plus backfill/downsampling and OpenMetrics exemplar support.
AI-written summary · about
chapters (38)
- 0:00Intro
- 0:40What is Prometheus
- 1:46Favorite Prometheus plugin
- 2:32Why Prometheus
- 3:15Operations
- 3:55Single monolithic application
- 5:01Storage backend
- 6:16Prometheus 10 problems
- 7:43Prometheus 10 test results
- 9:48Remote ReadWrite API
- 10:40Security
- 12:16Stability
- 13:21Release Cycle
- 13:43New Release Cycle
- 14:17Faster Prometheus
- 14:44Longterm storage
- 15:09Data islands
- 16:45Asset Database
- 17:00Isolation
- 17:37Beyond parameters
- 18:00Labels
- 18:31Open Metric
- 20:11Markers
- 20:59First adopters
- 21:10Change the world
- 21:23Codestep
- 21:35Hardware Vendors
- 21:54More Talks
- 22:05CPU Usage
- 23:25Prometheus Connect
- 25:44Prometheus Compression
- 28:09Prometheus QL
- 28:32Prometheus Refresh slowly
- 30:08Cortex and Thanos
- 30:55Local Storage
- 31:29Load Header
- 33:17Backfill
- 33:57Recording Rules
related talks
37mPrometheus Deep Dive
42mPrometheus Intro and Deep Dive
01h 19mIntro + Deep Dive: Prometheus
23mPrometheus – What it is, What is New, and What is Coming
33mPrometheus Project Intro
36mPrometheus - Intro, Deep Dive, And Open Q+A
33mIntro: Prometheus
37mJourney inside Prometheus: Architecture, updates, outlook