Chronix

Guide

Task-oriented documentation: how to install Chronix, write and query data, operate it, and connect the tools you already use.

If you want to know how it works rather than how to use it, the Internals section covers the algorithms and the reasoning behind them, and the Reference section documents each subsystem as it is implemented.

  • Getting Started Install Chronix as a Rust dependency or run the chronixd server. Write your first points and query them with the builder, SQL and PromQL.
  • Data Model Measurements, series, tags and fields in Chronix — how data is identified, why tag cardinality is the number that matters, and how time sharding, retention and deletes behave.
  • API Reference Every Chronix endpoint: REST write and query, PromQL, Prometheus remote write/read, OTLP metrics, gRPC, Arrow Flight SQL, and the embedded Rust API, each with a runnable example.
  • Analytics Forecasting (SES, Holt, Holt-Winters, ARIMA, SARIMA) and anomaly detection (Z-score, MAD, IQR, CUSUM, residual) inside the database, callable from SQL.
  • Operations Deploy and run Chronix: embedded and server modes, the full configuration reference, Prometheus metrics, Grafana dashboards, retention and rollups, backup and restore.
  • Performance Tuning Tune Chronix for your workload: memory budgets for constrained gateways, flush and compaction settings, object-store tiering, and the trade-offs behind each knob.
  • Security Authenticate with API keys, JWT/OIDC or mTLS; authorise with Cedar policies; scope every request to a namespace by credential; and keep a tamper-evident audit trail.
  • Client SDKs Talk to Chronix from Python, from any Arrow Flight SQL client, and from Telegraf, Prometheus and the OpenTelemetry Collector.
  • Grafana Integration Point Grafana at Chronix as a Prometheus data source, import the bundled dashboards, and query with PromQL or Flight SQL.
  • Configuration Reference Every Chronix and chronixd configuration key: defaults, units, and what each one trades away.
  • Examples Runnable examples covering the embedded API, SQL and PromQL queries, forecasting, anomaly detection, streaming, and the cold tier.
  • Cluster (frozen tier) The distributed Meta/Data/Query tier — frozen, kept compiling as reference until a post-1.0 thaw. Not part of the supported single-node product.