Concepts for Prometheus and Grafanas
Prometheus concepts describe how the system models and collects data. It stores metrics as time series identified by a name and labels, uses four core types (counter, gauge, histogram and summary), and pulls data by scraping HTTP endpoints defined in prometheus.yml.
Stage: Prometheus Basics · Part of the Prometheus and Grafana roadmap · Last reviewed 19 November 2026
What to learn in Concepts
- Explain a time series as a metric name plus label pairs
- Choose between counter, gauge, histogram and summary for a measurement
- Write a scrape_configs entry in prometheus.yml
- Start Prometheus and check target health on the Targets page
- Describe the pull model and when the Pushgateway is appropriate
Concepts interview questions
What is the difference between a counter and a gauge in Prometheus?
A counter only increases, or resets to zero on restart, and suits totals like requests served. A gauge can go up and down, and suits values like memory in use or queue length. Rates are taken from counters.
Why does Prometheus pull metrics instead of having apps push them?
Pulling lets Prometheus know which targets should exist, so a missing target shows as down. It simplifies configuration and lets you scrape a target by hand for debugging. Short-lived jobs are the main exception, using the Pushgateway.
Where Concepts fits in the Prometheus and Grafana roadmap
Concepts belongs to the Prometheus Basics stage (beginner level), which takes about 18 hours. Collect metrics.
Other units in this stage
Back to the full Prometheus and Grafana roadmap