Transform Data — modern ELT or ETL process?!
DEFINITION OF ETL & ELT
ETL (Extract — Transform — Load) is the process of combining data from multiple sources, transforming the data based on set of business rules and loading the data into a large, central repository called a data warehouse.
All three steps are done before the transformed data is finally loaded into data warehouse. In the other words, the data transformation occurs out of data warehouse.
On the other hand, ELT (Extract — Load — Transform) represents a reverse order of operations. The data is extracted and loaded into the data warehouse first and then transformed within the data warehouse.
MODERN ELT APPROACH OVER TRADITIONAL ETL
ETL process is a great solution when it comes to the expensive cost of data warehouse. In this context, data warehouse only stores transformed data which is previously processed in a staging server to save up the cost. However, nowadays the data size has been increased dramatically, staging server becomes the bottleneck. As a result, data analyst often have to wait for days before they can finally use the data to…
