on 01-26-2015 8:39 AM
Hi Team,
We are in a process to extract data from ECC to HANA and create views and push data to BW targets using ODP.
Basically we are extracting data from PS tables for example.
PROJ,PRPS,SKA1,T001,CEPC,CEPCT are master data tables.
COEP,COSP,COSS are the transaction data tables.
We would like to create attribute views for all the master data tables and for transaction data tables we would like to create calculation views rather than analytic views and push data to BW Targets using Operational data provisioning.
To extract from ECC - SLT
To push to BW Targets from HANA Views - ODP
Which is the best practice, is it calculation view or analytic view?
Please explain in detail, so that it would be great for our modeling.
Please explain pros/cons for both the views.
Best regards
venu
Venu,
I will suggest to go with calculation view with star join.
You can create calculation view as dimension for these tables: - PROJ,PRPS,SKA1,T001,CEPC,CEPCT
Then you can create calculation view with star join by including transaction table COEP,COSP,COSS as your base. Aggregate, join your data based on your requirement and then simply use above calc dim views in star join node.
Advantages: -
Link on how to use Calc view with star join : -
I am currently working on SP8 and these views works really well .
Regards,
Kulwinder
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi Everybody,
Thanks for your replies.
I would like to know how to handle WBS element and Project conversions from internal to External in our scenario.
Because if we use standard extractor we have PSPEX which is external WBS Element, as we extract directly from Tables (PROJ,PRPS) into HANA and then push to BW targets, where and how to handle this conversion.
Best regards
Venu
Hi Venu,
Based on experience, Olap engine is faster than calculation engines. There is no document available regarding this. So i cannot provide a detail explaination on this. If it is possible to achieve the view through analytical view then go for it. if not then go with calculation view.
Note that if there are too many calculated columns in your view go for calculation view since the calculation engine will come into picture.
Please go through below image for better understanding..
Hope this helps,
Cheers,
Safiyu
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi Venu,
Attribute views are basically dimensions that doesnt have any transaction data's in it.
If you have only master data's that has to be exposed, then go for attribute views.
If you have transaction data in you view then go for analytical view.
If you have complex calculations then go for calculation view.
Regards,
Safiyu
Hi Safiyu,
Thanks for your quick response.
Our current scenario is we have BW-IP running successfully from 3 years.
from Some of the regions there are data loads lead times, to avoid that we want to improve loading performance from ECC to BW via HANA.
and from there we have to push data to Planning cubes, that is our intention.
Please suggest, which is better way.
Best regards
Venu
Hi Patrick,
Analytical views make use of olap engine and olap engine is specifically designed to execute star schemas. basic master data and fact table join is nothing but a star schema, and olap engine processes them faster than calculation engine. I conducted a poc and analytical view was faster than calculation view. This is not applicable when your view has complex calculations. In that regard prefer calculation view.
Regards,
Safiyu
User | Count |
---|---|
85 | |
10 | |
10 | |
10 | |
7 | |
6 | |
6 | |
5 | |
4 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.