Skip to main content
Version: 1.2

User Guide

This guide covers running GreptimeDB: getting data in, querying it, processing it, and operating a deployment. For the data model and architecture behind these tasks, see Concepts. To install GreptimeDB and run a first query, start with the Quick Start.

Ingest Data​

Ingest Data maps each data source to the protocol it writes with and the guide that covers it. On most paths, tables and columns are created as data arrives; SQL, Flink, and Spark write into an existing table.

Logs and traces have their own guides, covering storage and querying beyond the write protocol:

  • Logs — parse and transform text logs with pipelines before they are stored.
  • Traces — store OTLP traces and query them with SQL or the Jaeger-compatible API.

Query Data​

  • SQL — queries across metrics, logs, and traces, including range queries, CTEs, joins, and views.
  • PromQL — metric queries through the Prometheus HTTP API, or TQL inside SQL.
  • Jaeger API — trace queries from Jaeger UI or Grafana.

The query overview compares the interfaces.

Process and Describe Data​

  • Flow — continuous aggregation from incoming rows into sink tables.
  • Data Index — inverted, skipping, and full-text indexes, and when each one pays off.
  • Update, Delete, and TTL — updating by overwrite, deleting, and expiring data.
  • Table Semantic Layer — greptime.semantic.* metadata recording the signal, source, and unit behind each table.
  • Vector Storage — the vector data type and similarity search.

Deploy and Operate​

  • Deployments & Administration — Kubernetes and bare-metal deployment, capacity planning, WAL modes, monitoring, disaster recovery, and Region operations.
  • Migrate to GreptimeDB — protocol, query, dashboard, and historical-data changes when moving from Prometheus, InfluxDB, Loki, Elasticsearch, or MySQL.
  • Time Zone — how the session time zone affects ingestion and query results.

Connect Other Tools​

  • Protocols — the wire protocols GreptimeDB speaks, and the boundaries of each one.
  • Integrations — Grafana, vmalert, Superset, Metabase, Flink, Spark, the MCP Server, and other tools that connect to GreptimeDB.