Glossary

Time-series database

A database purpose-built for data points indexed by time, optimized for fast writes and time-range queries.

Also called: TSDB

A time-series database is a database designed specifically to store and query data points indexed by time — sensor readings, application metrics, financial ticks, infrastructure logs — where nearly every write is a new timestamped value and nearly every query asks for a range or aggregation over time, such as an hourly average over the last week.

Time-series databases optimize for this pattern in ways general-purpose databases do not: extremely high write throughput for continuous streams of new points, automatic downsampling or retention policies that age out or compress old high-resolution data, and storage engines tuned for time-ordered, often columnar data with heavy compression, since consecutive readings tend to be similar. Query languages typically add built-in time-bucketing and interpolation functions on top of standard aggregation.

They are used for infrastructure and IoT monitoring, financial market data, and any application producing a continuous stream of measurements, often fed by a stream processing pipeline. A common pitfall is using a time-series database as a general-purpose store for data that is not actually time-ordered or time-queried, where its specialized optimizations offer no benefit and its more limited query flexibility becomes a liability compared to a general warehouse or Real-time OLAP system.

Last reviewed September 22, 2026

In the index now

Related terms

Real-time OLAP An analytical database that ingests streaming data continuously and answers aggregation queries with sub-second latency. In-memory database A database that keeps its primary copy of data in RAM instead of on disk, trading persistence for low latency. Columnar storage A file and database layout that stores each column's values together, instead of storing each row together. Stream processing Processing data continuously as individual events arrive, rather than waiting to accumulate a batch. Data warehouse A centralized, structured database optimized for fast historical reporting rather than day-to-day transactions. Window function A SQL calculation performed across a set of related rows without collapsing them into a single output row.

Related tools

Amazon Timestream AWS's managed time-series database, offered as a native serverless service or as managed InfluxDB. InfluxDB Purpose-built time-series database for metrics, sensor and IoT data, available self-hosted or as a managed cloud service. Prometheus Open-source monitoring system with a built-in time-series database, the de facto standard for Kubernetes and infrastructure metrics. QuestDB High-throughput open-source time-series database with SQL, built for fast ingestion of market, IoT and metrics data. TimescaleDB Open-source time-series extension for PostgreSQL, giving full SQL and joins alongside time-series performance. VictoriaMetrics Open-source, Prometheus-compatible time-series database and monitoring backend built for high cardinality at low cost.

Related guides

How to choose a time-series database Pick a time-series database by whether you need relational joins, what protocol your data already speaks, and how much cardinality you actually have. How to choose an industrial IoT analytics tool Historians, contextualization platforms, machine-health monitors and shop-floor apps solve different plant problems — map your job before comparing vendors. Time series analysis, explained How analysts model data ordered in time — the patterns forecasting methods look for, the pitfalls, and where the data actually lives.