Skip to Content
BW (SAP Business Warehouse)

Data Modeling


Can I combine a BW data model with a HANA data model?

Yes absolutely, and this is one of the great features of BW on HANA, that we can now provide a single platform for BW analysts as well as for SQL analysts, both consuming each other's data in a single data base.

Can I import a BW data model into the HANA data mart?

Yes, SAP supports cross-consumption of HANA models in BW as well as BW models in HANA. Currently as of HANA DB revision 37 and SAP notes 1703061 (please see ZIP file attachement), 1752384 and 1759172.

Please see these Video tutorials for more details:

It is very important to understand thou what this functionality is intended for. The linked document describes the positioning of this new functionality and what it should NOT be used for.

How can I report on a HANA information model directly in BW models? Is a consumption of HANA data models possible in the BW application?

The data stored in the HANA information models can be consumed either with a direct query on a Transient or a Virtual InfoProvider, or via CompositeProviders if data merge with BW data is required.

Does our team need technical training in order to use the HANA data modeling features in the HANA datamart?

In addition to the HANA data modeler, which is indeed intended for users extensive technical knowledge, SAP offers the HANA Information Composer for Business Users, which is Eclipse based.

Do I have access to the BW schema from HANA Studio?

With BW 7.30 SP8 several new features are provided on how to consume BW data in HANA outside the BW schema, and vice versa. See attached link for more info on BW 7.30 SP8: Mixed Scenarios

What is a BW workspace?

A BW Workspace is a highly flexible environment inside BW for business users to quickly react on new and changing business requirements. In this workspace users can build ad-hoc data models utilizing existing BW data and externally loaded data. BW Workspaces run completely in-memory.

There is a detailed document about several use cases where BW Workspaces would be a perfect fit. It includes details about the scenario, data model and the advantages of using Workspaces for those specific examples.

Does BW on HANA provide join operations, or only Unions as in the past, considering that InfoSets were limited in performance?

With BW on HANA you can still use InfoSets to create joins between InfoProviders. However, InfoSets execute the joins in the application layer and don’t push runtime down to the HANA DB. The recommended way to implement joins within BW on HANA is by using a CompositeProvider, because the joins are executed in the HANA DB directly. The CompositeProvider can also combine data from BW workspaces, APD  and also HANA real-time custom data marts via TransientProviders.

Exception: If temporal joins are required where the result set is depending on time-dependent master data, InfoSets have to be used since CompositeProviders can't provide that functionality yet.

Can I use the SAP BusinessObjects Universe instead of the HANA modeler in HANA Studio?

Universes can be used directly on tables in HANA and/or views of the HANA information model. It is recommended to model semantics inside HANA views in order to best leverage HANA's in-memory processing capabilities. Please see the links for details.

Do we still need aggregates with BW on HANA?

No, in BW on HANA we do not need any aggregates anymore.

Can I import or consume a HANA data model from a remote HANA system?

Yes, it is possible to consume or import HANA models that are not within the same system as you BW on HANA system and were created in another separate HANA instance. See the linked HowTo guide for more details.

Is it possible to find out how much database size objects in a specific namespace consume, e.g. to be able to distribute costs among different business departments?

There is two ways to do this. The first one is to leverage the database volume statistics gathered by BW's Technical Content in MultiProvider 0TCT_MC25.

Another option is to use the active/non-active data monitor which can be found in RSA1 --> Administration --> Monitors. This allows you to select a set of BW Infoproviders with namespaces or by InfoArea and then it displays the corresponding objects and all its DB tables. In this transaction you can also trigger a "load to memory" should the data of the tables not be fully loaded anyway. Then you simply sum up the column "total memory".

What are the best practices to deal with very large tables? What table sizes are recommended for large tables in BW on HANA?

Please make sure to review OSS note 2019973 which explains in detail all best practices around dealing with large data volume tables!

There is also a blog that describes in detail what table sizes of different InfoProviders are recommended and when partitioning makes sense.

Once BW data is exposed in HANA as native HANA views, what are the best practices to model natively in HANA?

Please refer to the Mixed Scenario FAQ page where you will find more information about this topic.

No comments