Prometheus.io

Functional Requirements

• FR1: Platform can ingest metrics (CPU, memory, latency, custom counters) from services.
• FR2: Users can query and visualize metrics on dashboards with filters, aggregations, and time ranges.
• FR3: Users can define alert rules with thresholds over time windows, e.g., “alert if P99 latency > 500ms for 5min”.
• FR4: Users can receive notifications when alerts fire, e.g., email, Slack.

Scale: 500K servers, 5M metrics per second, ~1GB/s raw ingestion.