Loading data before its transformation
ELT is very similar to ETL, but with a crucial difference: the order of the transform and load steps are inverted. This is very useful with big data in the cloud or when we do not have an ETL tool on-premises. This recipe will be much simpler than the previous one, as we'll implement ELT using a database, so no tools are involved here except for calling the ELT task.
It also relies on the previous recipe, Creating a simple ETL package, since we're going to use the SalesLT.CustomerFullName table data to implement the ELT pattern.
There are essentially two parts to this recipe:
- Extract and load data into our data lake. Here, we don't have a real data lake; we're using AdventureWorksLT on Azure to mimic the data lake concept.
- Transform the data inside the database using a simple SQL script. We're going to add the FullName column to the SalesLT.Customer table and update it using this script.
