From a single pod's CPU to your entire fleet's request rate, Nova ingests, stores, and visualizes metrics at any scale. Build dashboards that matter to your team, not generic templates that require 40 minutes of configuration to make useful.
High-cardinality metrics, tagged by user_id, trace_id, or deployment SHA, explode costs and degrade query performance in legacy observability tools. Nova's columnar storage engine handles millions of unique label combinations efficiently, so you can tag metrics the way your application actually works instead of stripping dimensions to control cost.
Start with a blank canvas, search for a metric, drag it onto the grid, and pick a chart type. No JSON config files, no YAML, no PromQL required unless you want it. Resize and rearrange panels by dragging. Nova's AI assistant can generate a complete service dashboard from a single description, "give me a dashboard for my payment service API."
Nova ships 200+ pre-built dashboards for every major infrastructure component: Kubernetes cluster health, PostgreSQL query performance, Redis memory utilization, NGINX request rates, AWS RDS, EKS node metrics, and more. Connect your data source and the dashboard appears, fully configured, within minutes of first data arriving.
Nova's anomaly detection engine runs continuously against every metric you ingest. On any chart, you can enable the anomaly overlay, a shaded band showing the expected range, with detected anomalies highlighted in red. Engineers reviewing dashboards during an investigation immediately see which deviations were genuine anomalies and which were within normal variance.
10M+ metrics per second, 50+ chart types, AI-generated dashboards, and anomaly overlays everywhere. 30-day retention included free.