Skip to content
Data

Time-series data

Data made of values recorded with a timestamp, in order, such as temperature readings every minute, which is the natural shape of most IoT telemetry.

Almost everything a sensor produces is time-series data: a stream of measurements, each tied to the moment it was taken. What matters is usually not one value but how values change, so queries ask for ranges, trends, averages over windows and comparisons between periods. The sampling rate sets how fine that picture is.

Time series have properties that generic databases handle poorly. Writes arrive constantly and in order, old data is read far less than recent data, and volume grows without bound unless a data retention policy trims it. Those properties are why a time-series database is a common choice for IoT storage and why downsampling old data is standard practice. The flow from device to storage is laid out in The IoT data pipeline.

Part of the IoT Glossary. Spotted an error or a missing term? Tell us.