Concepts

GreptimeDB is an observability database that unifies metrics, logs, and traces in a single engine. This section covers the core concepts you need to understand how GreptimeDB works and why it's designed this way.

Start here:

• Why GreptimeDB — The problem with three-pillar observability stacks, and how GreptimeDB solves it
• Data Model — How metrics, logs, and traces are represented as timestamped events with tags and fields
• Architecture — Compute-storage separation, stateless frontends, and how GreptimeDB scales

Deep dives:

• Observability 2.0 — Wide events, unified data model, and the evolution beyond three pillars
• Storage Location — Object storage, local disk, and multi-engine storage options
• Key Concepts — Tables, regions, time index, data types, views, and flows
• Common Questions — FAQ on updates, deletions, TTL, compression, high cardinality, and more

Further Reading

• Observability 2.0 and the Database for It — Our vision for the next generation of observability
• Unifying Logs and Metrics
• GreptimeDB Storage Engine Design