InfluxDB Time Series Database

InfluxDB is a free and open-source time series database (TSDB) purpose-built for storing, querying, and analyzing time-stamped data with high write and query performance. Developed by InfluxData (founded in 2013 by Paul Dix and Todd Persen, headquartered in San Francisco), InfluxDB is designed for monitoring, IoT sensor data, real-time analytics, and DevOps metrics. Key features: high-performance storage: optimized for high-volume time series ingestion, handling millions of data points per second. Data is stored in a columnar format (Time-Structured Merge Tree / TSM engine in v1, columnar storage in v3) with aggressive compression, achieving 5-20x compression ratios compared to raw data. Flux query language: InfluxDB v1 and v2 use Flux, a functional data scripting language designed for querying, processing, and transforming time series data. Flux supports data transformation pipelines, joins across data sources (SQL databases, CSV, HTTP), statistical functions, and anomaly detection. InfluxDB v3 (released 2024) uses SQL via Apache Arrow DataFusion, with InfluxQL as a legacy query language. Tags and fields: data points have a timestamp, tags (indexed metadata like host or region), and fields (measured values like CPU usage or temperature). Tags enable fast filtering; fields store the actual measurements. Retention policies: automatic data expiration and downsampling via continuous queries (v1) or tasks (v2). Downsampling reduces data granularity over time (e.g., store 1s data for 7 days, 1m data for 30 days, 1h data forever). Line protocol: compact, text-based ingestion format for high-throughput writes. Telegraf: InfluxData's data collection agent with 200+ input plugins for collecting metrics from databases, APIs, message queues, and sensors. Cross-platform (Go binary). MIT (v1-v2) / Apache-2.0 (v3).