influxdata/influxdb

InfluxDB is an open-source database built specifically for time series data, which is data where each record has a timestamp and represents a measurement at a point in time. Examples include server CPU usage sampled every second, temperature sensor readings from IoT devices, stock prices, application error rates, or network traffic counters. Unlike a general-purpose database, InfluxDB is optimized to ingest this kind of data at high speed and answer time-range queries very quickly. The version documented in this repository's main branch is InfluxDB 3 Core. It is built using a modern storage architecture: data is stored as Apache Parquet files, which is a columnar format that compresses well and supports fast analytical queries. Queries can be written in SQL or InfluxQL, the database's own time-series query language. The query engine is powered by Apache Arrow and DataFusion, which are open-source frameworks for fast in-process analytics. Data can be stored on local disk or in object storage like Amazon S3, Azure Blob, or Google Cloud Storage, giving a diskless deployment option. InfluxDB accepts data in line protocol format, which is a simple text format where each line describes one measurement with tags, fields, and a timestamp. It also provides HTTP and FlightSQL API endpoints for querying. You would use InfluxDB when your application generates time-stamped measurements that you need to store, query, and visualize, such as infrastructure monitoring, IoT sensor data collection, application performance tracking, or financial market data analysis. The tech stack is Rust for the core database engine, with Apache Arrow and DataFusion for query processing. The database can be run locally or in Docker, and the project is licensed under MIT and Apache 2.0.