So I assume you want to 'clean' the data before it enters the datawarehouse,
although you also contradict: 'a
large amount of data cleaning 9eletes, joins) _in_ the DWH'; do you want to
clean in between the Extract and Transform phase?
We do generate Extracts that don't need to be cleaned since the business
rules are coded inside Pro*C batches that do the extracts from the OLTP
system - less volume in the extracts. Also the business rules could be
applied in the batches that move the Atomic Data to the DataMarts.
If you could expand a little more on the problem you are dealing with, we
can perhaps give you more opinions.
"NorwoodThree" <norwoodthree_at_my-deja.com> wrote in message
news:ba03e2c.0302031352.188b5a79_at_posting.google.com...
> Is there any literature out there on the subject of Data Warehousing
> Best practices/ I've run into an issue where we are performing a
> large amount of data cleaning 9eletes, joins) in the DWH, and I'd like
> to propose that the data is cleaned with the business rules applied
> before it is loaded into the DWH. Opinions?
Received on Tue Feb 04 2003 - 15:29:37 CST
