Glossary · 7 · Product information systems and data roles
ETL and ELT
Also known as: ETL, ELT, Extract, transform, load, Extract, load, transform
ETL (extract, transform, load) and ELT (extract, load, transform) are two patterns for moving data from source systems into a target: ETL transforms the data before loading it, ELT loads it first and transforms it inside the target system.
- Data management
- System integration
In one sentence
Two patterns for moving data into a target: ETL transforms data before loading it, ELT loads it first and transforms it in the target.
Example
A nightly ETL job extracts released parts from PLM, maps their attributes and loads them into the spare parts catalog; a change made after the run reaches the catalog a day later.
Explained in context
Context cards connect this term with others to answer one question. Also in British English and German.
How it applies
- Timing: Batch ETL and ELT runs move data at set times. Between runs, the target holds the state of the last run; documents built from it carry that state.
- Transformation rules: Mappings, unit conversions and filters in the transform step change values. They belong in the data lineage of every value they touch.
- Pipelines: ETL and ELT are common building blocks of a data pipeline; event streaming and API calls are alternatives for flows that must be current.
ETL vs. ELT
ETL suits targets that should only receive cleaned, conformed data. ELT suits targets with enough processing power to keep raw data and transform it there, which keeps the raw copy available for later checks.
External references
- DAMA International, DAMA-DMBOK: https://www.dama.org/cpages/body-of-knowledge