The future of SAP Business WarehouseFebruary 21, 2021
Get your SAP data into Snowflake with Matillion using ODataSeptember 7, 2021
Despite great technological advances in SAP Business Warehouse (SAP BW, including BW/4HANA), enterprises still suffer from poor accessibility to SAP data. As SAP BW comes at a premium price, organisations are looking to see if other cloud data warehouse providers give better value for money. One of the reasons SAP BW was the de-facto choice was the ease of getting data out of SAP ERP, using SAP delivered ‘Business Content Extractors’. As Jan van Ansem explains in this blog, these extractors can now be used outside SAP BW, paving the way for creating a truly open data platform for both SAP and non-SAP data.
Why is it still so hard to integrate SAP data with ‘everything else’?
In a nutshell, this boils down to three main challenges:
1. People (skills and culture)
Typically, the SAP systems are looked after by people knowing everything about SAP data and technology, but very little about any other datasets and technologies. All other systems are looked after by teams who don’t want to touch SAP systems. Often, these teams are entrenched in their own bubble instead of collaborating on finding solutions for commonly shared problems.
Support for SAP BW in business applications is not as common and complete as for other data platforms. Popular BI & Analytics tools can usually connect to SAP BW, but users experience limitations which they do not have when connecting to other platforms. Business Applications outside the BI & Analytics (machine learning, robotics platforms, Industry/LoB specific applications) might not be able to connect to SAP BW at all.
3. License restrictions
SAP has a unique license agreement for SAP BW, which means certain features cannot be used – unless customers negotiate a different license agreement. Terms like ‘Open Hub’ and ‘Indirect Access Licensing’ still sends shivers down the spine of many brave data warehouse architect.
Are open cloud data platforms the answer to all the problems?
Platforms like Snowflake, Redshift, Microsoft Azure and Google Cloud Platform are great at delivering data warehouses at unlimited scale and great performance, but they come without the rich features for data warehousing SAP BW is equipped with. See my previous blog for more details on this.
Depending on the needs in your organisation, an open cloud data warehouse platform might be a better solution than SAP BW. If that is the case, you will want to know how to get data into the cloud, using the same extractors as SAP BW does.
How to get SAP data onto the cloud in a meaningful and efficient way?
The key here is ‘meaningful’. SAP BW has always benefited from the business content extractors, which are delivered by SAP. These extractors use complex logic to combine data from many tables into business entities. This means that when the data arrives in SAP BW, the data is almost immediately ready for consumption. If the tables were replicated without further context, BI developers would need to spend a huge amount of time creating data models to make the data usable for business users. Replicating tables is simple, replicating complex business logic less so. This has always been a stumbling block for getting data out of SAP and onto the cloud. Two frameworks have developed over the years and now reached a point where there are no more barriers to get SAP data onto a non-SAP platform:
- Operational Data Provisioning (ODP) Framework
The ODP framework provides a single interface to different types of extraction processes used in SAP analytics (including CDS views, HANA views and…. business content extractors). It is a subscriber-based model, based on modern message-based interface principles. (Here is a link to more info).
ODP is traditionally connected through RFC which reduces its usability as not many non-SAP applications can consume RFC connections. This is brings us to the second development:
- OData for SAP Products + ODP Based data extraction via OData
The SAP Gateway component offers an OData service for SAP data, which now also supports ODP-based data extraction. This means that you can connect the business content extractors to an OData connection, which in turn means that now any modern application (with an OData connector) can easily get data out of SAP.
This sounds somewhat technical, but it is not at all that complicated. Here is an overview of how the different components work together:
In the next blog (link here) I will take you through the process of getting data from SAP S/4HANA to Snowflake, using the standard business content extractors in an OData connection. It really is not difficult although there are some system requirements. Your SAP system needs to be ODP-enabled and supporting OData for SAP data framework version 2.0. With SAP Netweaver 7.5 you should be okay, for older versions you should check out these requirements.
The ability to use standard extractors to get data from SAP to the data warehouse of choice vastly reduces the implementation time and makes for robust, easy to maintain solutions. More importantly, this finally enables businesses to create a truly open data platform integrating all data, not just SAP data.