ETL is the classic data engineering workflow, and it still applies to IoT. A nightly job extracts the day’s readings from the time-series database, transforms them by converting units, joining with asset metadata and computing aggregates, and loads the result into reporting tables where business tools can query it. A variant, ELT, loads raw data into a data lake first and transforms inside it.
ETL suits reports and models that tolerate hours of delay. Its advantages are simplicity and reproducibility: a run can be repeated over the same period after a fix. Where reactions within seconds are needed, stream processing replaces or complements it, and the two usually share the same transformation logic.
Part of the IoT Glossary. Spotted an error or a missing term? Tell us.