What does ETL means ?

The acronym ELT means specifically Extract, transform and load. It became popular round 70's when organizations began using multiple data repositories, or databases, to store different types of business information.

It is really important part of business intelligence (BI) processes. The aim is getting data from different sources and put all of it in a data warehouse (DW). This process have to work automatically and that charge needs to run at least once a day, it depends on the kind of business.

Because of great amount of information usually trafficked across ETL and long processing time, in a DW the data are only information up to 1 day before. It is known as D-1.

The ETL is executed in steps, in the following:

1. Extract is the process of reading data from different sources. It could be a database, a file, other ETL processes or any other place or thing that has data interesting to be analysed. All data are sent raw to a place called Staging area which is a database where the data will be transformed.
2. Transform is the process of organizing the extracted data from the previous step of one way that make sense for business. That is the most important step of the process, because if it has failed, all project could be ruined.
3. Load is the process of putting all organized data in the transformations step into the DW which is a multidimensional model database

The process can be doing writing database code or using ETL tools.

There are some tools really useful to do ETL quickly and easily, it is following:

Talend, MS SSIS, IBM InfoShare DataStage, Oracle ODI and many others.

Just chose one and be happy!