ETL VS ELT

The idealogy for this E , T , L differes as needed in IT applications , so this ETL or ELT term used by IT vendors does not differentiate tools it differential architecture of application and how this ETL term is used . Since both do ETL and ELT does almost same work.

ETL (Extract, Transform and Load) - is software that transforms and migrates data on most platforms with or without source and target databases.

ELT (Extract, Load and Transform)- is software that transforms and migrates data in a database engine, often by generating SQL statements and procedures and moving data between tables. Its largly driven by RDBMS Vendors and they tended to be suitable for just one database platform.

Generally these terminology keeps coming , you may hear ETLT (Extract, Transform, Load and Transform) also in future .

Pros of ETL will be :
1. It can balance the workload with RDBMS
2. It can perform more complex operations .
3. Can scale with separate hardware.
4. Can handle Partitioning and parallelism independent of the data model, database layout, and source data model architecture.
5.Can process data in-stream, as it transfers from source to target
6. does not require co-location of data sets in order to do it's work.
7. captures huge amounts of metadata lineage today.
8. can run on SMP or MPP hardware

DATA QUALITY :- The ETL tools have a head start over ELT in terms of data quality integration . The row-by-row processing method of ETL works well with third party products such as data quality or business rule engines.

Pros for ELT will be :
1. It Leverages RDBMS engine hardware for scalability.
2. All data remains in RDBMS in all time.
3. The Disk I/O is usually optimized at the engine level for faster throughput.
4. All the capability of this is limited to RDBMS MPP platform.

The different part in both is architecture of application and technology used , Time line and other aspect depends on need of application and usage.