Hello,
The figures presented by Geetha are correct, but this is only the theoretical part (and applies very well to SQL sources like Google BigQuery, SAP HANA or any SQL DB you have connection with -> for SAP BW - no matter the version of your BW 7.3, 7.4, 7.5... #HANA or #NON HANA, the figures are highly different).
In reality, the end results of your data acquisition process from BW sources (BW Query) will get influenced by the number of dimensions from your query and the measures (Calculations; KF's; restricted KF; Aggregations and Exceptions Aggregations).
Be very careful that when you consume data from SAP BW because if you are expecting to have the same number of rows and columns as you see in Analysis Office, in SAC you will get a lot less. Even if changing the setup from RSADMIN in BW or you are increasing the SCC agent/ Tomcat server limitations, you will still have a dataset results less important than in Analysis Office.
Please be very careful with your expectation SAC <-> SAP BW (Data Acquisition) because you'll are going to be deceived at the end.
There are solutions and workarounds, but I find them very complex to use and not really friendly.
Tips to get around:
- Create queries in BW for that resume only to specif context -> avoid phrases "one query for all my needs";
- If you have errors/limitations- you can acquire each key figure in a separate sac query -> create a model with one measure and all the dimensions (once it is create) add a second measure into the model, create a new query for that new measure update it in SAC and continue like that until you finish;
- Create a model with a limited number of rows and initialise the model with the historical data from .CSV or Excel file exported from the BW query;
- Create complex calculations/restrictions and exception aggregations as much as possible in SAC (Model and Story) -> avoid using BW Queries with huge amount of calculations - complex formulas;
- Create as much as possible Public Dimensions with hierarchies (Key and Text) and use them into blank models in order to avoid a higher number of columns retrieval from BW;
- If your BW Systems allows it -> use HANA in order to exploit a higher number of rows and columns that you may be able get from BW;
- Imagine a mix between Analysis Office (for details - drill - slice and dice) and SAC (to have the "big picture") - think about the interoperability of these tow solutions;
I hope you have a better understanding of how to get along with BW data acquisition into SAC.
Regards,
Alex
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.