openGemini: A Distributed Time-Series Database
Shriira Press
A distributed time-series database built for the relentless write rates and millisecond queries of modern observability.
Welcome to openGemini: A Distributed Time-Series Database for Observability.
openGemini is an open-source, cloud-native time-series database designed to ingest, store, and analyze the enormous streams of metrics, logs, and traces that pour out of modern infrastructure, IoT fleets, and observability pipelines. It is a CNCF sandbox project, accepted in July 2024, and it sets out to solve the problems that grind ordinary databases to a halt at scale: millions of writes per second, time series numbering in the hundreds of millions, petabyte-scale retention, and the dreaded high-cardinality explosion that bloats indexes and devours memory. This book walks you through openGemini from the ground up. The opening chapters explain what a time-series workload really demands and why general-purpose databases struggle with it. From there we dissect openGemini's massively parallel architecture and its three core components — ts-sql, ts-store, and ts-meta — before going deep on the columnar storage engine, the indexing strategies that tame cardinality, and the way queries flow and push down through the cluster. Later chapters cover writing and querying data through InfluxDB-compatible and Prometheus interfaces, deploying and operating a cluster, and finally where openGemini fits in the broader cloud-native ecosystem.
This title is part of the ShriIra library and is free to read in full, right here — our small contribution to making world-class knowledge easy to reach.
A note on reading it: open the Contents menu at the top of the reader to jump between chapters, use the Aa menu to set a comfortable text size, theme (light, sepia, or night), and single- or two-page layout. Your place is saved automatically, so you can always pick up where you left off.
We hope it serves you well.
— Shriira Press