Explore the inner workings of Prometheus in this 36-minute deep dive conference talk featuring Prometheus maintainers Richard "RichiH" Hartmann from SpaceNet and Ben "SuperQ" Kochie from GitLab. Gain insights into the core functionalities, operational aspects, and future developments of this popular monitoring system. Discover topics such as storage backends, remote read/write APIs, security measures, release cycles, and potential improvements. Learn about data islands, asset databases, labels, and the Open Metric standard. Delve into discussions on CPU usage, compression techniques, PromQL, and integration with other tools like Cortex and Thanos. Understand the challenges faced by Prometheus and how the community is working to address them, making this talk invaluable for both beginners and experienced users of the platform.
Overview
Syllabus
Intro
What is Prometheus
Favorite Prometheus plugin
Why Prometheus
Operations
Single monolithic application
Storage backend
Prometheus 10 problems
Prometheus 10 test results
Remote ReadWrite API
Security
Stability
Release Cycle
New Release Cycle
Faster Prometheus
Longterm storage
Data islands
Storage
Asset Database
Isolation
Beyond parameters
Labels
Open Metric
Markers
First adopters
Change the world
Codestep
Hardware Vendors
More Talks
CPU Usage
Prometheus Connect
Prometheus Compression
Prometheus QL
Prometheus Refresh slowly
Cortex and Thanos
Local Storage
Load Header
Backfill
Recording Rules
Taught by
CNCF [Cloud Native Computing Foundation]
