What is the current SAP Data Warehouse Strategy?

Share post via

With SAP BW/4HANA, SAP HANA SQL Data Warehousing, and SAP Datasphere, SAP offers three different data warehousesolutions. The strategic focus on public cloud products based on the Business Technology Platform (BTP).

Further developments and future innovations are only to be expected in this area. Conversely, this means that all other solutions will no longer be developed intensively. Datasphere is SAP's strategic application-driven cloud data warehouse solution and will therefore be the focus of future investments and innovations by SAP.

Graphic - On-Premise vs. Private Cloud vs. Public Cloud
Fig. 1: On-Premise vs. Private Cloud vs. Public Cloud | isr.de
Graphic - On-Premise vs. Private Cloud vs. Public Cloud
Fig. 1: On-Premise vs. Private Cloud vs. Public Cloud | isr.de

🛈

For more information about Datasphereand a good overview, please visit our Themepage and in our overview flyer.

SAP does not actively promote the HANA SQL Data Warehouse approach as a DWH product. However, the HANA Cloud is strategically positioned as the basis for the Datasphere, other software products (e.g., SAP Analytics Cloud), and individual software development. A HANA SQL Data Warehouse can be operated locally, in the private cloud, or in the HANA Cloud. This means that the HANA SQL approach to data warehouse development is consistent with SAP's strategy. The HANA SQL Data Warehouse approach complements the SAP Data Warehouse portfolio with a "native driven" approach to data warehouse development and stands for a high degree of openness and flexibility in modeling. Compared to the ready-made DWH application SAP BW, it is more in line with the tradition of open data warehouse approaches. Tools and methods are used that are also used in other data warehouse environments such as Microsoft Azure, AWS Redshift, or Snowflake. The core of the tools used are part of HANA native software development. These include:
  • Power Designer 
  • XSA 
  • Web IDE 
  • Smart Data Integration (SDI) 
These SAP HANA-specific tools are supplemented by tools that are standard in software development, including those that enable agile, DevOps-oriented development: 
  • Git 
  • Jenkins / Bamboo 
  • tracking tools 
The methods used are also very similar to those used in other data warehouse environments. The central basis is conceptual and physical data models, which are modeled and implemented according to common modeling standards (3NF, Data Vault 2.0) and then supplied accordingly via ETL processes.  The HANA SQL DWH approach is characterized by the following features:
  • Model-driven development process  
  • Agility and DevOps orientation 
  • Use of software development standards 
  • Automation in data warehouse development 
  • Cloud readiness 

🛈

Further information about Datasphere and a good overview can be found on our Themepage.
BW/4HANA is SAP's classic plug-and-play BI solution, which has been complementing the SAP portfolio as a DWH application for years, especially in relation to enterprise resource planning. Thanks to this long-standing relationship, BW/HANA continues to demonstrate its strengths in SAP-based system landscapes and has proven itself as a data warehouse solution. SAP positions BW/4 HANA as a strategic solution for an application-driven data warehouse that can be operated in your own data center (on premise) or the private cloud. Support is guaranteed for the long term (2040) and BW/4 HANA is part of the Rise with SAP strategy, which provides a certain degree of investment security. However, due to SAP's focus on the (public) cloud, no major innovations are to be expected in BW/4 HANA. Further developments will primarily focus on hybrid architectures (interaction between BW/4 HANA and the Datasphere and Analytics Cloud). With the BW Bridge, however, SAP offers the option of transferring at least parts of BW architectures to the Datasphere, thus facilitating the long-term transition to the cloud.

🛈

Further information about Datasphere and a good overview can be found on our Themepage or in our flyer "Data Warehousing with SAP BW/4HANA."

How is SAP's Datasphere positioned?

The analytical landscapes at companies are becoming increasingly complex. In addition to an enterprise data warehouse (more than one solution is conceivable for special topics), data lakes are often found in architectures. The EDWH is managed by IT and is subject to central governance. In the specialist departments, there are increasingly more employees with analytical skills who want to independently expand data models and calculate key figures, especially for short-term issues. Therefore, there are local departmental solutions (e.g., SQL Server, Tableau Data Prep, Power Query, etc.) to close this gap. Until now, these components have often functioned as loosely coupled units that are (too) poorly integrated.

Graphic - Positioning SAP Datasphere1
Fig. 2: Positioning of SAP Datasphere | isr.de
SAP positions Datasphere as a bridge between EDWH and local departmental solutions by integrating both worlds into a single technical platform. In this respect, Datasphere is strategically positioned as
  • Enterprise Data Warehouse Setting up an EDWH and making it available to the organization, e.g., as a centrally governed space 
  • Self-service data preparation for specialist departments User-friendly modeling for departments to independently expand the IT-governed EDWH—unlike isolated solutions, however, it is integrated into the same system environment as the EDWH. 
With the help of the data layer, the datasphere should be able to integrate a variety of systems: 
  • Cloud-based software solutions 
  • On-premise systems These can be operational systems, but also existing data warehouse solutions.  
  • Data lake(s) Either directly integrated as a HANA data lake or through integration of external data lake solutions with Datasphere 
  • External data and sharing data with third parties The Data Marketplace is a marketplace for data. The basic idea is that it allows publicly available data to be quickly and easily integrated into your own data models and analyses. At the same time, the Sharing Cockpit allows you to make your own data available to other definable Datasphere tenants (e.g., your own customers or local subsidiaries). 
Graphic - Positioning SAP Datasphere
Fig. 3: Positioning of SAP Datasphere| isr.de

In contrast, the business layer, based on the data models of the data layer, can be used to provide cross-system, uniform semantics for analytics tools. The datasphere then functions as a central data virtualization layer for reporting.

Is Datasphere already suitable for building a complex enterprise data warehouse?

Datasphere is a new product. We would therefore like to explore the question of whether Datasphere is already suitable for building a fully-fledged enterprise data warehouse. One focus is on corporate landscapes that require the construction and operation of a large and complex EDWH. In this article, we would like to examine the issue from the perspective of SAP BW and HANA SQL customers. SAP BW/4 HANA is a product that has been around for over 20 years and has been refined and optimized for SAP source systems over the years. In addition to BI content, its modeling capabilities have also been optimized for SAP structures. It is therefore not surprising that Datasphere cannot (yet) offer the same range of functions. There are limitations, particularly in terms of modeling options (e.g., compounded keys, currency conversion). BW/4 also offers functional advantages in the areas of authorization control and lifecycle management. We could mention a few more details here, but the bottom line remains the same: Datasphere still lacks some functionalities compared to BW/4 HANA. When comparing the HANA SQL approach, it should first be noted that the approaches to DWH development differ greatly from one another. While an application-driven approach specifies certain development and modeling paths, the native-driven approach allows for greater openness in development. The architect decides how the DWH is modeled (e.g., Data Vault) and does not have to conform to the conventions of an application-driven approach. At the same time, the integration of Git repositories and build tools allows for a high degree of openness for DevOps (including CI/CD)-oriented procedures. All of this enables a high degree of agility in development. In terms of authorization control, lifecycle management, and monitoring, the requirements are comparable to those of a BW/4-oriented approach. With regard to these aspects, there are still more or less significant limitations, meaning that even compared to the HANA SQL approach, there is still some way to go in terms of enterprise readiness in large, complex environments. We therefore conclude that Datasphere is not yet capable of building a complex enterprise data warehouse. At the same time, the SAP roadmap shows that a great deal is being invested in this area. In this respect, it will be necessary to observe closely and assess for yourself when Datasphere has reached a level of maturity that is sufficient for your needs. In the explanations provided so far, we have only briefly touched on the topic of enterprise readiness as an enterprise data warehouse. In our view, the following aspects are relevant for an enterprise data warehouse. We will therefore address the topic of enterprise readiness in more detail in a separate article.
Fig. 4: Structure of a complex enterprise data warehouse | isr.de
Fig. 4: Structure of a complex enterprise data warehouse | isr.de

However, the assessment of the EDWH use case does not mean that Datasphere is generally not ready for use in companies. The evaluation must always be made in the context of the use case. Customers with a relatively simple architecture and corresponding use cases, as well as those who want user-friendly modeling, will potentially already be able to map their requirements with Datasphere today. In our view, Datasphere can also complement the Analytics Cloud and existing DWH solutions very well to provide self-service data preparation options for specialist departments. Opportunities may therefore lie in hybrid architectures, where Datasphere can bring its strengths to bear.

What opportunities do hybrid architectures offer in conjunction with the datasphere?

In the preceding question, we discussed that, in our view, the datasphere is not yet ready to map a complex EDWH. However, we already see advantages in combining the datasphere with existing architectures.

Datasphere as a self-service platform 

The Datasphere is very well suited for simple and user-friendly data modeling. The possibility of SQL script and data flows from Data Intelligence also opens up several ways to implement logic. In the example below, the coordinated and harmonized standard data models are provided in a governed space in the EDWH. In companies, we often encounter more than one solution that performs DWH tasks. These other possible distributed solutions can also be provided in a governed space. In the Datasphere, departments can access the quality-assured area to expand the data models as desired (for example, with external data from the Data Marketplace). The advantage of this architecture is that the departments gain or retain flexibility in an integrated architecture where a certain level of governance can be maintained. Looking ahead, we can also well imagine the Datasphere functioning as a virtual semantic layer across all systems. This would then enable a uniform technical view of different data sources and offer it to front-end applications.
Graphic - Datasphere as a self-service platform
Fig. 5: Datasphere as a self-service platform | isr.de

SAP BW/4 HANA and Datasphere

BW/4HANA exhibits weaknesses in the area of self-service data preparation. BW Workspaces have never been an optimal option. We identify numerous opportunities in a hybrid architecture combining BW/4HANA with Datasphere. From a technical standpoint, various options exist for integrating these solutions.
Fig. 6: Hybrid Scenarios | isr.de

1. Remote Tables / Database Connection

In this scenario, BW/4 HANA is connected in the data layer and used as a remote source. This allows access to data in the BW system. On the one hand, this scenario is very quick and easy to implement. On the other hand, the Datasphere has no knowledge of the semantics of the BW/4 HANA system. If you only follow this hybrid approach, you have to manually recreate the semantics in the Datasphere. This is the simplest form of integration for quickly making data models available in the Datasphere so that specialist departments can carry out enhancements.

2. BW/4 HANA Model Transfer

In contrast to a simple database connection, the Model Transfer Connection links the BW Query. Datasphere not only reads the necessary database connection but also extracts the semantics, creating the required objects in the Business Layer. This type of connection offers the significant advantage of eliminating the need for manual Business Layer modeling. Conversely, the first option provides greater flexibility in modeling if a business entity needs to be defined differently from BW/4HANA, for example, by extending BW data for an entity with external data.

3. BW Bridge

The BW Bridge is a BW/4 HANA environment created in the Datasphere environment. The bridge provides the Datasphere with BW data models in a "bridge space" – so there is no native BW integration into the Datasphere modeling objects. From a BW perspective, the BW Bridge offers limited functionality. For example, only ODP data sources can be connected, not all provider types are supported, and the "analytical layer" (e.g., query) is missing. Details on the possibilities and limitations are described in Note 3117800.

The BW Bridge does not integrate existing BW/4 HANA systems into the Datasphere. However, with the BW Bridge, BW customers have the option of transferring or migrating their BW systems to the Datasphere environment. Theoretically, this makes it possible to migrate existing systems. However, the functional scope of the bridge is very limited from a BW perspective, meaning that many things would have to be recreated manually in the native "core" area of the Datasphere. In the future, the bridge will offer the same two integration functions (remote tables + model transfer) as described above.

From Datasphere's perspective, however, BW Bridge opens up a number of possibilities. ODP-enabled SAP source systems can be connected to the Datasphere via the Bridge. The Bridge can act as a kind of ingestion layer here. At the same time, BW/4 HANA content is available, so the Bridge should be set up and filled with data relatively quickly. The content's InfoObjects also act as accelerators for building SAP-based data models in the Datasphere.

Graphic - SAP BW Bridge
Fig. 7: SAP BW Bridge | isr.de

SAP HANA SQL DWH, BTP, and Datasphere

HANA SQL DWH, in BTP or on-premise, has a very high level of integration with Datasphere, which is only natural since Datasphere is ultimately based on HANA Cloud. There are basically two options for integration: via the Open SQL schema or via HDI containers. Integration is easy in both directions.

The exchange via HDI containers is particularly interesting. This makes it easy to integrate structures and content into the Datasphere that were developed on the HANA Cloud with Business Application Studio or on HANA on premise with WebIDE. This also works in the other direction, making it easy to deploy from the Datasphere to the HANA Cloud or HANA on premise.

Graphic - Business Technology Platform
Fig. 8: Business Technology Platform | isr.de

This makes it easy to set up a scenario in which, for example, the enterprise data warehouse is built in BTP (HANA Cloud native), where complete enterprise readiness is already in place. The LoB-oriented part, with its strengths in self-service, is then located in the Datasphere.

Further information can be found here:

SAP BTP Showcase – Access the SAP HANA Cloud database underneath SAP Datasphere | SAP Blogs

SAP Datasphere integrated with SAP SQL data warehousing – accessing Datasphere consumable models as Source | SAP Blogs

What should companies facing DWH modernization do?

What do customers who want to modernize their data warehouse do? One option would be to look for alternatives in the non-SAP area. This is also something we notice in customer discussions. But is this necessary? Can I build my enterprise data warehouse with SAP? We will attempt to provide an assessment based on three initial scenarios: SAP BW customers who are considering modernizing their data warehouse by introducing SAP BW/4 HANA are faced with the question of whether this is still viable for the future. Yes, the introduction of BW/4 HANA is viable for the future and the investment is not a dead end. We have come to this assessment for the following reasons:
  1. BW/4 HANA is a highly sophisticated data warehouse product that is being used very successfully. You can be sure that the product is very well suited for building an enterprise data warehouse. There will be few functions and features that the product still lacks.
  2. At the same time, there is a roadmap for BW/4 HANA and a commitment from SAP that BW/4 HANA will continue to exist until at least 2040. That's 18(!) years, which is a (very) long-term confirmation of investment security.
  3. With BW Bridge, SAP has created a (public) cloud perspective for BW customers in 2021. The bridge is a (streamlined) BW/4 HANA development environment that runs in the Datasphere environment. It gives customers the opportunity to migrate at least parts of their BW architectures to Datasphere. We will explain what BW Bridge actually is and what possibilities it offers in a separate article.
Against this backdrop, BW customers should keep a close eye on the market and the further development of Datasphere, but should not allow themselves to be unsettled. For companies modernizing their DWH with SAP HANA SQL Data Warehouse, the question of future viability also arises. The technology and methodology used are fully compatible with HANA Cloud. This means that a HANA SQL Data Warehouse is already cloud-ready per se. Customers can choose whether they want to deploy their native data warehouse on-premises, in the private cloud, or, in the future, in the HANA Cloud. Since Datasphere is based on HANA Cloud, the HANA SQL approach can be easily integrated with Datasphere to expand a HANA native EDW with more self-service options, for example. The HANA SQL approach can also be considered secure in terms of investment, as unlike BW/4, it is not a product, but a selection of specific tools and appropriate methods.
  • The tools are SAP tools, as used in SAP HANA Application Development. They are therefore subject to the respective SAP product strategy.
  • The corresponding methods are strictly based on general DWH development standards, meaning that the HANA SQL DWH approach differs only slightly from the structure of other SQL-based DWHs, such as those in a Microsoft environment.
Overall, HANA SQL DWH can also be considered a safe investment, making it SAP-compliant and very future-proof. HANA Cloud native EDWH offers companies that want to set up an EDWH in the cloud today but do not want to do so in a non-SAP environment a viable option. As mentioned above, the HANA SQL DWH approach can also be used in the cloud, as it is already cloud-ready. This variant is not officially proclaimed by SAP, but it exists and works, and can be considered just as secure an investment as the HANA SQL approach, which is widely used. This approach is very similar to the development of EDWHs in an AWS, Microsoft Azure, or Snowflake environment and thus represents an alternative if you want to remain cloud-based in an SAP environment. Companies that previously had a non-SAP data warehouse may have deliberately decided against SAP BW in the past. In these cases, we have often observed that a more open DWH approach is desired than the application-driven BW approach allows. We recommend that customers who have consciously decided against a BW-based approach also consider the HANA SQL approach or the HANA Cloud native approach. Native DWH development is comparable to non-SAP DWH development approaches. With HANA Cloud, SAP has provided a highly scalable and high-performance platform that, in our view, is in direct competition with Snowflake & Co. If, on the other hand, companies have no "reservations" about an application-driven "all-in-one" approach or want it, we recommend including BW/4 HANA in the evaluation as well.

Conclusion

SAP BW/4 HANA and HANA SQL are no longer available, but BW/4 HANA, HANA SQL, and Datasphere are here to stay!

This is how we could briefly summarize the conclusion of our article. It is clear that Datasphere is SAP's primary strategic DWH product. There should be no illusions about this. The central element of the strategy is the HANA Cloud. This means that the HANA SQL DWH approach is also part of the strategic portfolio.

Does this mean the end of BW/4 HANA? Not (yet). The life cycle of BW/4 HANA will end in 2040 at the earliest. With BW Bridge, there is also a cloud perspective for parts of BW/4 HANA developments. At the same time, no further major investments will be made in SAP BW.

outlook

This blog post is part of a series that addresses current issues relating to SAP's revised DWH strategy (e.g., Is Datasphere enterprise-ready as an enterprise data warehouse? What is the BW Bridge, and what migration options are available?).
ISR Employee Image

Christopher Kampmann
Head of Business Unit
Data & Analytics
christopher.kampmann@isr.de
+49 (0) 151 422 05 448

Xing Linkedin

About ISR

Since 1993, we have been operating as IT consultants for Data Analytics and Document Logistics, focusing on data management and process automation.
We provide comprehensive support, from strategic IT consulting to specific implementations and solutions, all the way to IT operations, within the framework of holistic Enterprise Information Management (EIM).
ISR is part of the CENIT EIM Group.

Visit us virtually on these channels:

Xing Linkedin Facebook-f Instagram Youtube
News Categories
News Archive

Latest Publications

Upcoming ISR Events

[tribe_events_list limit=”3″]