Integration between SAC and BPC Standard Model

[SarauShen]

In this blog, I will introduce BPC and SAC integration solution. BPC is the strategic planning solution in on-premise area and SAC is the strategic planning in Cloud area. Hence hybrid solution of SAC and BPC is combination of two best worlds.

BPC Standard Model

New workflow with some enhancements based on new DA framework is shipped in wave 2019.01. Details

Integration Types

Import/Export

The BPC to SAC import/export was developed to have a smooth way to replicate Financial planning models or parts of them to SAC. BPC plays the role of corporate planning and SAC plays the role of line-of-business (LOB) planning. These financial planning models are typically not that huge in BPC.

Typical workflow is like:

1. To prepare import/export workflow, IT should setup the connection in SAC.


2. LOB acquires Corporate planning target data. The data acquired is usually part of corporate planning data via filtering in import process. You can setup a schedule to do the import from BPC model to SAC model on a regular basis.


3. LOB does planning process internally. LOB can add more master data for local planning purpose. SAC advanced planning capabilities(spreading, allocation, powerful table, version management…) could be fully leveraged.


4. Please note only the fact data is imported to SAC. The calculated data from member formula is not imported.


5. LOB submits its final planning data back to corporate. The planning functionalities in BPC can be leveraged, such as script logic.

Writeback

A writeback model is introduced for semi-live integration between BPC and SAC. In order to have semi-live integration, the master data should be consistent between SAC and BPC.

By leveraging the delta region based data synchronization, a near live experience could be achieved. Customers could still leverage the planning capabilities, such as logic scripts, from BPC. In the meantime, customers could also enjoy flexible planning capabilities, such as spreading, distributing, predictive forecast and powerful visualization features from SAC. SAC now could serve as an alternative rich client on top of BPC models.









There are some things worth mentioning.

1. All the dimensions created in SAC will be private dimensions.


2. All master data(except formula attribute) will be read-only in SAC since they should be synced from BPC.


3. When DAC (Data Access Control) is enabled on any dimension in SAC, the read & write access could still be maintained in SAC. The DAP (Data Access Profile) setting in BPC could not be synced automatically yet to SAC, so you will need to set up your DAC setting in SAC independently.


4. Currency dimension will not be imported into SAC. Currency conversion need to be configured inside SAC separately if necessary.


5. SAC Input Task is NOT supported in writeback model.


6. Right after SAC user publishes the data from story to BPC, SAC will retrieve the latest transaction data from BPC, and the calculated data generated by BPC member formula will be retrieved along with the non-calculated data by default. Depending on the filter range defined for the BPC writeback model, the volume of fact data generated by BPC member formula can be large and the calculated data retrieval process can be performance concerned. Alternatively, if the SAC user does not want the calculated data generated by BPC member formula to be involved during data refresh, there's an IMG parameter named with "SKIP_GET_FORMULA_DATA" on BPC side. The parameter is on BPC model level. The BPC administration user may go to transaction SIMGH to add the parameter to corresponding BPC model and set the value of parameter as "X". After the parameter has been set, BPC will not return the calculated data in SAC writeback scenario.


7. Be very cautious to activate BPC data on the BW DataStore Objects (advanced) or compress BPC data on the BW Infoproviders at BPC side. Otherwise the time stamp information can be removed from the activated or compressed requests and you might not be able to sync delta data from corresponding BPC model. Refer to note https://launchpad.support.sap.com/#/notes/2915785


8. Writeback model is created based on the selected BPC model, and timestamp of latest refresh from BPC model is recorded in SAC model. Hence the writeback model can't be redirected to another BPC model because timestamps of different BPC models are different.


9. Transaction data can't be deleted via writeback


10. Edit import/export is not supported

Refer to blog: https://blogs.sap.com/2017/08/03/easy-data-communication-with-deeper-integrated-hybrid-solution-of-s...

What are replicated with restriction

• Member length of Currency dimension is 3 in SAC, hence the members with more than 3 characters are truncated.


• Attribute length is 127 in SAC, hence the attributes with more than 127 characters are truncated.


• SAC time dimension supports 12 months, hence the months other than 12 months are ignored.


• In subsequent import, update of member properties/new master data in BPC can be synced to SAC. Deleted members can't be synced to SAC (hence can't be deleted in SAC in subsequent import).

• In subsequent import, update(incl. zero out) of transaction data/new transaction data in BPC can be synced to SAC. Deleted transaction data (after compress zero) behaves differently in Basic Model and Writeback Model.

What are not replicated

• Data access profile can't be replicated automatically from BPC to SAC.The data access profile is designed differently in BPC and SAC, hence it's not feasible to replicate data access profile. BPC DAP can be set via hierarchy formula, member filter and attribute filter. SAC DAP currently can be set via member filter. Apart from it, it's technical challenge to get the SAC users from BPC users to assign data access profile to users.


• Member formula can't be replicated automatically from BPC to SAC. Member formula has different syntax in BPC and SAC. It's not feasible to convert the member formula from BPC to SAC. The BPC formulas that produced the calculated members are imported into an unused FORMULA column of the account dimension. You must recreate the formulas in the proper syntax for SAP Analytics Cloud in the  Formula column.


• Calculations defined in EPM plugin in AO can't be replicated to SAC.


• Script Logic can't be replicated. The Script Logic in BPC and Advanced Formula in SAC can be used separately in the planning process.

What are recommended

• BPC and SAC integration is targeted to replicate planning model, whose data volume is usually not huge.


• Detailed data could be aggregated in BPC and replicated to SAC to avoid mass data in import/export to gain a better performance.


• Replicating data on-demand instead full data from BPC to SAC via filtering in import process to gain a better performance.

What are not recommended

These scenarios are not recommended in BPC/SAC integration:

• Import BPC YTD model to SAC (period storage) . Converting from YTD to Periodic is quite time-consuming. And BPC YTD model is usually used as Consolidation model, which is not planning scenario. In rare case that BPC YTD model has to be imported to SAC model, and user would like to see YTD data in SAC story, he/she could create restricted measure with YTD value.

• Large amount of transaction/master data is not recommended for the time being. The data volume supported depends on memory setting in BPC system and transaction data distribution. The more equally transaction data is distributed on master data, the larger data volume supported. There is a plan to improve the scalability.