Question:
Hi everybody,
I wish to know how and why using the custom table as a datasource is preferable to linking BW to third party systems. Actually, we have five third party systems provide data besides cost center data and financial data from R/3. We are debating whether to get the data after transformation to a custom table in R/3 to be used as a generic data source. On the other hand, there is an idea to directly link BW to excel csv files. Can someone tell me which is preferable? I have my own doubt about using excel as it has limitations to number of rows while we expect the line items to exceed more than 65000 rows per period.
Thanks.
Som
Answer:
You have CSV files, so just use Flat FIle upload to get the data into BW from your legacy systems. If you are not using/needing the data in R/3 there is no need to put it into R/3 to pull it into BW.
Answer:
Thanks SBobcat for the reply. Yes, we will be pulling financial data from R/3 and master data like cost center. If so, is custom table is the answer? What DB connect has to do with this?
Thanks for the instant reply.
Som
Answer:
DB Connect is also an option instead of flat files. Search this forum on that and flat files and you will find a wealth of information discussing these different options.
Answer:
DBconnect is only valid for certain third party
csv files don't have a limitation to 65000 rows, only if you want to see it in excel ; csv is only a format of file
no need to put it in r3 if not used in R3
_________________
_
There are only 10 types of people in the world :
those who understand binary and those who don't.
Answer:
I agree with both CHC and SBobcat. We wanted to use DB Connect with an AS/400 but DB COnnect to AS/400 only works if the BW is running on Windows application servers ( windows + server != that what we have an want), so we might have to use *.csv instead of this as the data won't be used in R/3.
If you use a new file every load, there's no problem about delta management too, always load the new data via a full upload and it works pretty fine. A disadvantage of *.csv is that you have to define the transfer structure completely manually and it has to be exactly according to the layout of the according flat file.
_________________
Juergen Limbach