VictoriaMetrics

VictoriaMetrics is an open-source time-series database and monitoring backend designed for high ingestion throughput, efficient storage, and PromQL compatibility. It is commonly used as a backend for Prometheus-compatible monitoring stacks and is suitable for large-scale deployments.

The project offers two deployment modes: a single-node setup using the vmsingle binary for small environments and testing, and a cluster mode that uses separate components for ingestion, query, and storage. In cluster mode, multiple vminsert instances ingest data, vmselect handles queries, and vmstorage stores the time-series data. The cluster can be scaled horizontally to meet throughput and retention requirements, with data partitioning and replication across storage nodes.

Key features include a Prometheus-compatible API and PromQL, support for remote_write and remote_read, Grafana compatibility, and mechanisms for data retention, auto-compaction, and compression to reduce storage usage. The design emphasizes low resource consumption and high ingestion performance.

Use cases include large-scale monitoring and long-term storage of metrics, particularly in Kubernetes environments. VictoriaMetrics integrates with standard Prometheus scrapers and Grafana dashboards and can serve as a backend for dashboards and alerting in sizable deployments.

Licensing and development: VictoriaMetrics is distributed as open-source software, with an enterprise edition offering additional features and commercial support. The project is maintained by VictoriaMetrics Ltd and an active community of contributors.