Technical Guide: Overview for Migration to SAP BW/4HANA

Key Concept

As the pace of change increases, the ability to create analytic solutions quickly and effectively can mean the difference between an organization performing or outperforming. The concept of agile data warehousing solves many of the thorny problems typically associated with data warehouse development—most notably high costs, low user adoption, ever-changing business requirements, and the inability to rapidly adapt as business conditions change. SAP BW/4HANA is technically prepared for fast deployments and rapid delivery of complex business analytics scenarios.

SAP BW/4HANA brings a new opportunity for agile enterprise data warehousing (EDW) modeling and fast delivery of business analytics’ requirements for organizations seeking solutions in reporting, planning, dashboarding, Big Data, and predictive and prescriptive analytics. SAP is working on combining SAP HANA and SAP BW into one product, and, today, SAP BW/4HANA is the recommended technology for an EDW platform by SAP.

While SAP BW/4HANA leverages the capabilities of native SAP HANA on large datasets, it also does the following:

• Simplifies and speeds up data modeling by using reusable master data with texts, hierarchies, and navigational attributes
• Improves jobs scheduling and monitoring via process chains
• Offers a simplified approach to modeling and multi-temperature architecture
• Provides non-cumulative modeling for inventory management and headcount
• Makes available business content for SAP originating models
• Offers an analytics manager for faster development of complex queries

SAP BW/4HANA now has an increasingly agnostic focus across the enterprise and is more open to receiving and processing non-SAP and unstructured social media data.  The SAP Business Planning and Consolidation 10.1 embedded model is supported with the SAP BW/4HANA Starter add-on.

What Makes SAP BW/4HANA Different

SAP BW/4HANA is not just a new version of SAP BW. SAP BW/4HANA is only available on the SAP HANA platform, and is not part of SAP NetWeaver—it has been completely rewritten and optimized for SAP HANA. All future innovations are associated with the new product and not with the older series of SAP BW 7.x on SAP HANA. In SAP BW/4HANA, traditional SAP business intelligence (BI) business content is no longer supported. There is new SAP HANA-optimized content instead that leverages the Operational Data Provisioning Framework-based extractors, advanced DataStore objects (DSOs), and CompositeProviders.

If you look at the big picture, including SAP S/4HANA (a recent ERP application from SAP), SAP BW/4HANA is completely independent from and is not a prerequisite for SAP S/4HANA, and vice versa. SAP S/4HANA provides operational reporting for current data from a single SAP application. SAP BW/4HANA delivers a modern data-warehousing environment that allows organizations to report and plan on current, historical, and external data sources from many SAP and non-SAP applications.

Simplified Modeling, Increased Agility

The next version of Layered Scalable Architecture (LSA++) standards are designed to provide simplified modeling and architecture. Data models become more open, dynamic, and flexible, removing borders between data warehouse layers and reducing the number of persisted data stores.

In LSA++, users are no longer required to create multiple layers of aggregated DSOs and InfoCubes. Most of the key performance indicators (KPIs) can be designed using SAP BW queries and CompositeProviders that logically connect objects together in unions or joins. CompositeProviders and SAP BW queries can be exposed to native SAP HANA models in complex predictive and data-mining scenarios.

It is fair to say that, in SAP BW/4HANA, the propagation/staging layer (e.g., advanced DSOs) and virtual data marts (e.g., CompositeProviders) can be used for most reporting and planning modeling requirements. In some specific scenarios, though, corporate memory (advanced DSOs), architected data marts (advanced DSOs), and agile data marts (SAP BW workspaces), can be added to the architecture as well (Figure 1).

Most of the existing data flows in SAP BW can—and should—be simplified in SAP BW/4HANA by leveraging LSA++ principles and by its ability to access data directly in the staging/propagation layer without compromising performance (Figure 2). A lot of complexity in the old flows becomes redundant and must be completely removed when implementing the simplified LSA++ architecture.

New Source Systems and Business Content

There are fewer source system types in SAP BW/4HANA and, at the same time, more data-provisioning opportunities for real-time data integration via the ODP–SAP Landscape Transformation (SLT) replication server or SAP HANA Smart Data Integration, direct/virtual data access (via SAP HANA Smart Data Integration), or data loads (via SAP HANA-optimized BI content and extractors [ODP]). The ODP source system replaces the SAP source system. All Business Content data sources (extractors) that are released for ODP are supported (Figure 3).

New SAP BW/4HANA Business Content was released in December 2016. It combines the enterprise data warehouse capabilities of SAP BW?4HANA with explorative and interactive real-time analytics using the SAP HANA in-memory database. It includes a preconfigured set of application models for selected business areas and industries. Delivered data models follow the recommendations for the LSA++ optimized for SAP BW?4HANA. The data models contain queries, CompositeProviders, DSOs (advanced), InfoSources, key figures, characteristics, and transformations for SAP R/3, S/4HANA, and other selected applications. Business Content for SAP BW?4HANA also includes data flows for master-data provisioning of InfoObjects, which are used in these data models.

There are two new distinct content add-ons:

• The SAP BW/4HANA Basis Content add-on (BW4CONTB) provides (besides other objects) a set of almost 3,000 InfoObjects (from the former BI_CONT) and a newly defined set of variables.
• The SAP BW/4HANA Content add-on (BW4CONT) provides data flows and transformations for master-data provisioning of the InfoObjects and transactional data flows and models (e.g., advanced DSOs and SAP HANA CompositeProviders for different business areas—for example, sales and distribution, financials and material management, and industry [such as retail and utilities]).

For more information, follow this link: https://help.sap.com/viewer/p/SAP_BW/4HANA_CONTENT.

SAP HANA-Optimized Models, and Faster Loads and Query Execution

In the SAP HANA-optimized models, there is much fewer persistency objects and, therefore, fewer data loads are required. Business-logic complexity, lookups, and data-integration rules are normally hidden under CompositeProviders, SAP BW queries, and SAP HANA views. SAP BW/4HANA operates only with SAP HANA-optimized objects; therefore, data loads and query executions run substantially faster.

Due to the in-memory nature of the SAP BW/4HANA database, no aggregates or roll-up processes are required—there is no need to design performance-specific objects and there are fewer database indexes. All these advantages lead to faster data loading and data processing.

Smaller Data Footprint

Figure 2 shows an example of a data-flow optimization for SAP BW/4HANA by reducing several data layers and persistency objects. Specifically, in the new data flow there are two advanced DSOs at the staging layer and a CompositeProvider. All kinds of data aggregations, slicing, and calculations are usually done in the virtual layer by leveraging CompositeProviders, SAP BW queries, or SAP HANA views. This optimization, together with the compression of SAP HANA data, leads to a substantially smaller data footprint and smaller SAP HANA database size requirements. When planning a migration to SAP BW/4HANA, the same sizing process used for SAP BW powered by SAP HANA applies to the migration to SAP BW/4HANA. (For more information about this, see SAP Note 2296290.)

Field-Based Modeling: When Do You Still Need InfoObjects?

Objects in SAP BW?4HANA can be modeled with InfoObjects or with fields. Field-based modeling can be especially useful when integrating non-SAP data. Modeling InfoObjects requires more effort, but offers several advantages. When deciding whether to model with fields or with InfoObjects, you should consider the modeling flexibility offered by fields versus the enhanced functions and performance that come with the InfoObjects

Field-based modeling allows faster integration of external data sources as it no longer requires that InfoObjects be created one by one. Connecting and consuming external tables in a CompositeProvider or an SAP BW query becomes easy via SAP HANA Smart Data Integration. Field properties and aggregation rules can be configured directly in the CompositeProviders.

There are a few restrictions, however, where InfoObjects are still required. For example, in the embedded version of SAP Business Planning and Consolidation, all the components of direct-update advanced DSOs must be defined as InfoObjects. Another example is non-cumulative models (e.g. stock or inventory reporting), where InfoObjects are required. If  you need to leverage attributes or hierarchies, InfoObjects are required as well.

SAP HANA-Optimized Transformations

Complex business logic is usually defined in transformations. When transformations are required, it is crucial to make them SAP HANA-optimized for better performance. When a transformation is activated, the system checks whether it can be performed in SAP HANA. If the transformation has an ABAP Managed Database Procedure, it must be performed in SAP HANA.

Click the check icon  to the right of the transformation’s name field to see if the transformation can be performed in SAP HANA (Figure 4). The system then attempts to create the transformation in SAP HANA. If this is successful, the transformation is flagged with Can be performed in SAP HANA. If the check is not successful, you can find out why by viewing the log.

SAP HANA Analysis Process and Predictive Analytics

You can use an SAP HANA analysis process (commonly referred to as HAP) to analyze data from certain perspectives—for example, to calculate ABC classes or to find scoring information and outliers.

With the SAP HANA analysis process, SAP BW/4HANA users can use the functions of the SAP HANA database and combine these with the functions in the SAP BW system. An SAP HANA analysis process is always made up of exactly one data source, a function for data analysis, and a data target. An SAP HANA analysis process can also be used as the source for another SAP HANA analysis process.

Multi-Temperature Architecture and Partitioning

Partitioning or semantic groups may be useful when modeling advanced DSOs with large amounts of data (Figure 5). SAP BW/4HANA has a limit of 2 billion records per advanced DSO; however, for better performance during reporting SAP recommends that the number of records in one advanced DSO or partition be limited to a few hundred million (e.g., 500M maximum).

SAP BW/4HANA has a flexible dynamic-tiering capability for advanced DSOs on the object or partition level, leveraging multi-temperature architecture.

New Data Flows

New modeling capabilities combine advanced DSOs, transformations, CompositeProviders, Data Transfer Processes (DTPs), and source systems into new Data Flows (Figure 6). This is a neat way to group modeling objects together by functional area and to maintain enhancements.

SAP HANA Studio vs. SAP’s Graphical User Interface (GUI)

In SAP BW/4HANA, the SAP GUI—for the most part—is replaced with SAP BW modeling tools that provide an Eclipse-based, integrated modeling environment for the management and maintenance of SAP BW?4HANA metadata objects. These tools integrate with ABAP Development Tools as well as with SAP HANA modeling and the consumption of SAP HANA views in SAP BW?4HANA metadata objects, like Open ODS views or CompositeProviders. The SAP BW Modeling perspective provides you with predefined views and configuration for your SAP BW modeling tasks organized in projects.

Although most of the old transaction codes are still available in SAP BW/4HANA, transaction code RSA1 is no longer used the same way as in classic SAP BW, and developers are encouraged to use the new Eclipse-based interface including new Data Flow modeling capabilities.

Leverage BW Queries, Native SQL, and Mixed Scenarios

With the modeling functionality available in either SAP BW?4HANA or SAP HANA, mixed modeling makes it possible to access SAP BW?4HANA data (e.g., queries, advanced DSOs, InfoObjects, and CompositeProviders) from any schema in the SAP HANA database, and to access data from any schema in the SAP HANA database (e.g., tables and views) from SAP BW?4HANA (Figure 7). You can thus create scenarios where data, which is modeled in the SAP BW?4HANA system, is merged with data modeled in SAP HANA.

The BEx Query Designer tool is no longer available in SAP BW/4HANA. Old BEx Queries can be converted to SAP BW queries in the SAP HANA studio BW perspective. Moreover, SAP BW queries automatically generate SAP HANA calculation views that can be consumed in SAP HANA SQL-based applications and predictive analytics (Figure 8).

Here is a link to the page describing how SAP HANA views are generated for SAP BW queries:

https://help.sap.com/viewer/107a6e8a38b74ede94c833ca3b7b6f51/1.0.2/en-US/c470990abfce4796aacf7a41a27fedcc.html

SAP BW queries in SAP HANA studio have the same capabilities or more compared to the old BEx Query Designer tool. In addition to the older SAP BW on SAP HANA version advantages and SAP HANA optimizations, execution of queries is pushed down to the SAP HANA server (including exception aggregation). Also, execution of SAP BusinessObjects Business Planning and Consolidation embedded planning functions (e.g., disaggregation) is pushed down to the SAP HANA server.

Key Considerations When Making the Decision to Migrate

Agility of data models, cloud infrastructure, and self-service-user capabilities are key contributors to successful business analytics solutions. SAP BW/4HANA makes an important step forward in terms of improving EDW’s agility and flexibility. Complex modeling solutions (e.g., reporting, dashboarding, planning, and predictive) on large datasets can be designed in days instead of weeks, and introducing changes to business logic or data models becomes a much easier process going forward.

SAP BW/4HANA advances development of a modern data warehouse to address new types and locations of data whether from SAP applications or any other source. SAP BW/4HANA also supports the trend of analyzing data in its original location to minimize movement and duplication of data, including data lakes and data in motion. It delivers real-time flexibility and overcomes the challenge of data silos. The addition of sophisticated data-temperature management and compression ensures that data is tuned for both results and cost.

A set of new features and capabilities in SAP BW/4HANA leads to improved agility. This is due to its simplified architecture (SAP HANA-optimized queries exposed to SQL on top of CompositeProviders connecting objects from various data layers), simplified modeling capabilities (fewer modeling objects, new data flows, field-based modeling), easier access to variety of data sources either directly (SAP Smart Data Interface) or in real time (SLT and ODP), and SAP HANA-optimized Business Content that improve performance.

Agility and simplified architecture lead to new possibilities in analytics, creating more efficient models, smaller data footprints, and, consequently, lower TCO. Faster response time to changing business requirements is another benefit. Building mockups and proofs of concept for various reporting, planning, and predictive scenarios becomes a much easier task and is faster to implement.

Cloud readiness allows new ideas to be prototyped and put into operation in days, not months, with minimal upfront investment. Out-of-the-box deployment to SAP HANA Enterprise Cloud, Amazon Web Services (AWS), or another provider make the cloud-deployment option a great choice to start with SAP BW/4HANA. SAP Cloud Appliance Library (CAL) has an SAP BW/4HANA offering for both trial and subscription usage (for more information about SAP CAL, follow this link: https://cal.sap.com).

You should also leverage a range of self-service business analytics tools on top of SAP BW/4HANA. They can be deployed at a relatively low cost, either on premise or in the cloud. A range of modern front-end capabilities from SAP include SAP Business Objects Analysis, SAP Lumira, Design Studio, BusinessObjects Predictive Analytics, and BusinessObjects Cloud. SAP front-end tools are normally better integrated with SAP BW/4HANA in terms of hierarchies’ integration, unit/currency conversions, planning data-write-back capabilities (Integrated Planning and Business Planning and Consolidation embedded), and report-to-report interfaces. However, companies also have other options, and sometimes prefer non-SAP self-service BI tools. Popular BI vendors and tools (Microsoft Power BI, Tableu, and Qlick are just some examples) can connect to SAP HANA data warehouses, offering companies a multitude of choices when selecting front-end user capabilities.

Sergei Peleshuk

Sergei Peleshuk has more than 15 years of experience implementing BI technologies for global clients in retail, distribution, fast-moving consumer goods (FMCG), oil, and gas industries. He has helped clients to design robust BI reporting and planning capabilities, leading them through all project phases: from analysis of requirements to building BI roadmaps, technical architecture, and efficient BI teams. Sergei is an expert in modern BI tools and technologies available on the market, including SAP Business Warehouse (SAP BW), SAP HANA, BusinessObjects, and SAP Lumira. Sergei maintains a business intelligence portal at www.biportal.org.

Sergei will be presenting at the upcoming SAPinsider HANA 2017 conference, June 14-16, 2017, in Amsterdam. For information on the event, click here.

You may contact the author at peleshuk@biportal.org.

If you have comments about this article or publication, or would like to submit an article idea, please contact the editor.