Prometheus is an open-source time-series database and monitoring system. It collects metrics by scraping HTTP /metrics endpoints on a configurable interval (pull model), storing labeled time-series data locally. Target discovery uses static configuration or dynamic service discovery integrations (Kubernetes, Consul, AWS EC2, and others).
PromQL is the query language for Prometheus data. It operates on labeled time-series and supports aggregation, filtering, and mathematical operations across arbitrary combinations of label dimensions. Recording rules precompute expensive PromQL expressions as new time-series for dashboard and alert efficiency.
Alertmanager receives firing alerts from Prometheus, applies grouping rules to collapse related alerts, inhibition rules to suppress downstream alerts when an upstream alert fires, and silence management, then routes notifications to Slack, PagerDuty, email, or webhooks. Remote write forwards metrics to external long-term storage backends (Thanos, Cortex, Mimir, VictoriaMetrics), extending retention beyond the local TSDB. Prometheus is a CNCF graduated project under Apache 2.0.
