In doubt, use common sense.
Proverb
There are no fixed guidelines on when to use ETL or ELT. It is normal that both are used in many enterprise companies and normal.
It is clear that you can design a process and it is working beautifully for you, 100% OK.
But what is happening when you extrapolate this problem to a scale that your process it is taking longer and longer time because your database has become a MONSTER…
In this post, we do a quick analysis on each method with Pros and Cons of each methodology.
ETL (Extract-Transform-Load)
ETL is more typical than ELT because of the existing GUI Tools that help the user to set the process outside of the data source.
ELT (Extract-Load-Transform)
ETL
ELT
In Dimensigon Data Platform, you can integrate all your data sources without the need of loading the data and therefore consume a big amount of resources in the Cloud (Which it is expensive). Through this functionality, you can run your reports and queries cross-system (Almost any database) and including HDFS or flat-files.
DIMENSIGON DATA PLATFORM