Insightful data management has always been a key priority for businesses, but, at least initially, the analysis was more focused on structured data for longer-term planning and strategic decision-making. With advancements in technology and digital disruption, such as the rise of Internet of Things (IoT), multimedia, and social media, the focus has shifted to unstructured and streamed data. The current database-warehouse system does not have the capability to support so many different types of query processing and different data locations; hence, companies are looking for new data-warehousing solutions. The move to digital organizations has led companies to focus on prescriptive analytics. There have always been three type of analytics. Descriptive refers to the past to summarize what has happened. Predictive talks about the future, making predictions based on the past data. And prescriptive analytics provides different scenarios, along with their impact on the business to help answer the question “What should be done?”. SAP BW/4HANA supports companies in doing all three types of analytics by combining data sources together to create insights. It helps companies to combine their own database with social media platforms to generate such data, and then to send personalized offers and promotions to their customers. SAP BW/4HANA also allows companies to:

• Collect and choose structured data, unstructured data (Twitter, Facebook, and so on), and streamed data (such as the IoT) from multiple sources.
• Choose a data-management option to leverage this new technology, such as vertical and graph databases, Hadoop, and so on, along with traditional data-query processing
• Have the capability to build advanced analytical models, such as predictive modeling, to help with decision-making and to grow their company’s profits.
• Have real-time analytic environments and faster project turnarounds.
• Provide business users with self-service BI tools that give them direct access to transactional data to create models.

The Architecture of SAP BW/4HANA

SAP BW/4HANA is built on top of the in-memory SAP HANA platform and merges the functionality of SAP HANA with SAP Business Warehouse (Figure 1). SAP BW/4HANA supports the standard warehouse use cases, such as operational reporting and historical analysis, and allows users to perform research, advanced customer analysis, and sentimental analysis, as well as develop new algorithms and create behavioral models in real time. It also simplifies the integration of real-time data with historical data to enhance decision-making. SAP BW/4HANA is based on four principles:

1. Simplicity
2. Openness
3. Modern interface
4. High performance

I discuss the main architecture points that differentiate SAP BW/4HANA from other data warehouse solutions in the following sections.

Simplified Source System and Data Integration

SAP BW/4HANA allows you to analyze data from multiple locations, such as data lakes (e.g., Hadoop) or real-time IoT data from sensors. It offers the flexibility to analyze the data at the original location without moving the data to an SAP BW/4HANA database. This reduces redundancy of data and saves time. With SAP BW/4HANA, the number of source systems has been reduced from 10 to three (Figure 2), which, in turn, has simplified data integration (Figure 3). The reduced SAP BW/4HANA consists of the following source systems (Figure 2):

• SAP Operation Data Provisioning (ODP): This is used for all the SAP back-end systems and SAP Landscape Transformations (SLT).
• SAP HANA source system: This is used for all the databases and file connectivity. SAP HANA leverages SAP enterprise information management (EIM) for new data provisioning.

SAP BW/4HANA offers simplified data integration with SAP HANA smart data integration (Figure 3). With SAP HANA, developers get the flexibility to use SAP HANA smart data integration to fetch the data to SAP BW/4HANA.

SAP HANA smart data integration and SAP HANA smart data quality enhances, cleanses, and transforms data to make it more accurate and useful. With the speed offered by SAP HANA, SAP HANA smart data integration, and SAP HANA smart data quality, you can connect with any source, provision and cleanse data, and load data into SAP HANA on-premise or in the cloud. It supports all the types of data delivery, including:

• Federated
• Batch
• Real time

SAP BW/4HANA also includes SAP HANA smart data access and cross-database access with multitenant database containers for fetching the data. SAP HANA smart data access enables access to remote data-access sources without copying the data to SAP BW/4HANA. Smart data access supports a variety of data sources, such as Teradata, Sybase IQ, and Sybase ASE. SAP BW/4HANA allows the collection of IoT data using the SAP HANA streaming option.

Streamlined Data Life-Cycle Management

Most companies have lots of data. Companies are accumulating this quantity of data to avoid being left behind. As a result, companies are becoming data rich, but have poor insight into this deluge of data. As per research by Forrester, companies today are still using BI and Big Data in silos, which is not providing correct insights. In addition, data also has a limited lifespan and—at some point—it becomes outdated and irrelevant. Companies invest a lot of money in storing these enormous banks of data, which requires them to identify strategies for gaining insight into this data, at less cost. SAP BW/4HANA use a temperature-based model to ensure that the right data is available when needed and that resources are used efficiently (Figure 4). Data managed by BW can be classified into four different categories:

1. Hot data: Data that is frequently accessed (e.g., for reporting, data manipulation, and so forth) and is available on the SAP HANA platform.
2. Warm data: Data that is accessed less frequently and can be stored in a low-cost memory option. SAP BW/4HANA uses the concept of SAP HANA dynamic tiering to store warm data. In classical BW, the classification as warm can only be done at the object level, which means that the entire persistency is either warm or not. In BW/4HANA, the Advanced DataStore object (ASDO)-selected range partitions can be classified as warm data to allow a more granular classification.
3. Cold data: Data that is rarely accessed. The data-archiving process is used to move data slices from the online database to the near-line database by using a near-line service. The data is moved by means of a secondary database connection from the ABAP server to the near-line database. Currently only SAP IQ NLS and SAP Hadoop NLS are supported for near-line storage for BW/4HANA. The near-line storage process has been more simplified for the BW/4HANA.
4. Raw data: This is unstructured and streamed data. SAP Hadoop is used to store unstructured data and real-time IoT streamed data.

The report access for the near-line storage and raw data on SAP Hadoop is done using SAP HANA smart data access. Refer to SAP Notes 2316757 and 2317200 for more detailed information on the extension and slave for hot and warm data.

Native SQL Access

When you activate SAP BW/4HANA objects, such as ASDOs, CompositeProviders, and so forth, SAP HANA calculation views are automatically generated with the same structure as the BW/4HANA objects (Figure 5). All the front-end tools, such as SAP Lumira and SAP BusinessObjects Design Studio, can access the generated SAP HANA views for reporting. The generated views can then be used for predictive analysis or other scenarios for reporting.

Simplified Data Structure

SAP BW/4HANA reduced the number of objects from 10 to four (Figure 6). That helps to reduce the complexity for designing, implementing, and modifying the data-warehouse environment. This also helps to expedite the development process for creating new scenarios, such as:

• CompositeProviders: A CompositeProvider is an InfoProvider that combines data from several analytical indexes or from other InfoProviders (by join or union) and makes this data available for reporting and analysis.
• Open Operational Data Store (ODS) views: Open ODS views enable you to define BW data models for objects like database tables, database views, or SAP BW data sources (for direct access). These data models allow flexible integration without having to create InfoObjects.
• ADSOs: The ADSO is the only object you need for persistence. It replaces all previous types of DataStore Objects, persistent staging areas (PSAs), InfoCubes, and Hybrid Providers.
• InfoObject: InfoObjects are the main building blocks of the data model and can be divided into key figures or characteristics.

Performance

As SAP BW/4HANA is built on top of SAP HANA, it leverages all the features of SAP HANA, as shown in Figure 7. SAP HANA supports all the built-in innovations in SAP HANA, such as the aggregation feature and the ability to calculate on the fly, without any data redundancy. In addition, SAP BW/4HANA offers:

• Algorithm push down: SAP BW/4HANA can push-down operations or calculations to SAP HANA. Once the operations are moved to an in-memory database, they can be performed much faster and more efficiently.
• Advanced analytics: SAP BW/4HANA supports all the advanced analytics’ feature of SAP HANA. SAP HANA provides a lot of analytical functions in the Application Function Library (AFL), Text Analysis, Predictive Analysis, and Machine Learning.

SAP BW/4HANA and SAP BW S/4HANA

SAP BW/4HANA and SAP S/4HANA are completely independent of each other. Thus, SAP BW/4HANA can be used without SAP S/4HANA. Companies can use SAP S/4HANA for operational reporting and SAP BW/4HANA for historical analysis. SAP S/4HANA and SAP BW/4HANA have a single source of truth (e.g., both use the same data). SAP BW/4HANA further eliminates the data duplication and expensive data movement from the Online Transaction Processing (OLTP) database to the Online Analytical Processing (OLAP) database for reporting.

SAP BW/4HANA Deployment Options

SAP BW/4HANA is available both on premise and in the cloud. The cloud version enables companies to set up their infrastructures quickly. The cloud version of SAP BW/4HANA can be deployed on the SAP HANA Enterprises Cloud, Amazon Web Services, and Microsoft Azure.