A high-performance, high-reliability, high-scalability, and low-cost distributed time series database which focus on the storage and analysis of massive time series data


High Cardinality

A new high cardinality storage engine solves problems such as excessive index memory usage and low read and write performance

Data Analysis

Armed with extensive data analysis algorithms, openGemini supports real-time anomaly detection and forecasting reducing the workloads on OPS engineers.

Low O&M Costs

Scure and stable, simple architecture, quick deployment and no third-party dependencies

Data Compression

Data is stored in column format, and different data types use dedicated data compression algorithms. The data compression ratio is as high as 15:1 or higher

Application Scenarios


openGemini: Batch Query Optimization

By Zhi Chen / 2023-11-24

This article introduces what is batch query and the performance problems caused by serial execution of batch query. openGemini makes full use of the powerful concurrent processing capability of Go language to carry out concurrent transformation of serial query, and proposes an adaptive parallel query task scheduler to solve the performance problems of serial query.

openGemini Optimization Example: Query a Plan Template

By Hanxue Li / 2023-11-16

Database performance tuning is a process that requires patience and perseverance. It is necessary to continuously analyze and optimize, find out potential performance problems, and take appropriate measures to optimize. Only through continuous efforts and practice can we really improve the performance of the database system and provide better support for the business. This article presents an example of how to optimize openGemini based on flame graph analysis, and also describes how to add a query template to help readers.

openGemini: Full-Text Index Parsing

By Jie Li / 2023-11-10

This article introduces the function of openGemini full-text index, including the basic principle, index creation, full-text query and filtering. Compared with the traditional full-text index, the CLV dynamic segmentation algorithm adopted by openGemini has great advantages in the memory resource consumption and matching efficiency of inverted index.

The Application Scenarios and Constraints of the Multistage Downsampling Function

By Guanglin Cong / 2023-11-01

This article mainly introduces the openGemini multilevel downsampling function, including application scenarios, task creation, view and deletion, and usage constraints. Multilevel downsampling can greatly reduce data storage costs and system costs, but it does not retain the original data details, you must be clear when using.

Join Us


Gemini is open source.Star our GitHub repo,follow us on Twitter,and join our developer community on Slack


We look forward to working with more enterprises, universities and developers to jointly promote technological innovation

Privacy StatementLegal Statement


Docker Repository




Copyright @2023 OpenGemini-All Rights Reserved.