ETL Extraction


I am running an ETL process that uses timestamp on the source to determine the changes. We have to use this method as the source data is located on a SQL server 2005 machine.
This method will increment the timestamp when any of the rows are changed. I am only extracting data from a small number of columns within the table.
For example I need the data on col1, col2, col3, col4 only. If the data changes only in col5, the ETL process extracts the rows those rows as well. It does not do anything with it once extracted but it seems like an unnecessary action.

Is it possible to extract only the rows in which changes to the columns I want to extract have changed and ignore the rows where other columns have changed?

Many thanks
Who is Participating?
ste5anConnect With a Mentor Senior DeveloperCommented:
Well, the extract process must of course extract all values, before he can determine which columns/rows have changed. Otherwise you need an change indicator in the source data and you *must* trust the source data for correct change tracking.

So in the extract phase this is pretty normal. In the transform and load steps you may skip unchanged data.

Just for curiosity: Do you mean ROWVERSION, when saying TIMESTAMP?
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.