Describing SAP Business Data Cloud and its Design Principles

SAP Business Data Cloud (SAP BDC) is a fully managed Software as a Service (SaaS) solution that unifies and governs data from SAP and non-SAP applications to power advanced analytics and AI. SAP BDC includes many new innovations that are required to power a data platform that is able to handle the demands of AI now and in the future.

SAP BDC was developed to meet the challenges faced by organizations who want to build a modern data fabric to support current and future AI powered applications, but are held back by their current data platforms.

SAP BDC provides the data layer that bridges the applications and AI. SAP BDC provides a data platform to manage all SAP and non-SAP data and is used to build advanced analytical applications and AI.

SAP BDC is not a single software product. SAP BDC is a platform that includes multiple software components that work together to provide all the services required of an AI-ready data platform. Customers choose which of the listed SAP BDC components they want to use depending on their use case. They can start with a simple SAP BDC landscape and grow this as their requirements evolve.

To fully understand why SAP BDC was needed, it is important to understand the data fabric approach.

A data fabric is a combination of data architecture and dedicated software solutions that centralize, connect, manage, and govern data across different systems and applications.

SAP BDC is an evolution of the data fabric journey that began with SAP Datasphere. SAP BDC does not replace SAP Datasphere. In fact, SAP Datasphere is an essential, central component of SAP BDC. But many other software components were needed to support a complete data fabric. SAP BDC includes all the components required to support a complete data fabric.

You can learn more about data fabric here: What is a Data Fabric?

Let's learn how SAP BDC addresses each of the data management challenges that organizations are currently facing.

• Extraction from SAP applications can be complex and costly.

  • Instead of customers building data pipelines from their SAP applications, SAP takes care of the extraction and combining of customers' data from all of their SAP applications to the central data store of SAP BDC. SAP make the data available so it is business-ready. This means the data can be used right away in analytical applications without the need for preparation.

• Multiple SAP applications provide the same object but with different definitions.

  • SAP BDC unifies the business objects from different SAP applications. For example, if SAP S/4HANA provides Material and Concur provides Product they are harmonized into a single definition so that the business user does not need to know, or care about the source of the data.

• Business semantics are lost during extraction from SAP applications

  • SAP BDC not only extracts the raw business data but also all of the rich semantics so that the business meaning is not lost. This means that customers do not have to rebuild the lost semantics after extraction of data.

• Incomplete data sets

  • SAP BDC provides data extraction from all SAP applications so you can build analytics on all your SAP data with no missing data sets.

• Building analytics take too long

  • SAP BDC provides ready-to-use analytics that can be used immediately by business users.

• Our data can't be trusted to build reliable AI

  • SAP takes care of the provisioning of all data from customers' SAP applications into a secure and trusted data store ready to use in AI development projects.

• Storing data at scale is expensive

  • SAP BDC stores data at scale by leveraging a lakehouse architecture on hyperscalers for cost-efficient, high-performance storage of data in files

• Data governance is difficult

  • SAP BDC provides a secure and centralized data governance model that ensures an organization knows where all there data is stored, who has access to data and for what reasons

SAP BDC includes analytical content for the business user to get started immediately. The analytical content is delivered by SAP as Intelligent Applications. Intelligent applications are data-rich dashboards that use real-time data and are powered by AI to support queries using natural language. For example, a user can simply ask "Which are the top selling products in region A?" or more complex questions such as "Why have my profits decreased in Central America compared to last year?". The SAP BDC intelligent application generates the results and offers additional insights that the business user might have missed that might trigger new lines of inquiry.

SAP BDC has been developed to support a modern data product economy.

A data product economy treats data as a product that has been cleaned, refined and packaged. Individual data products are business-ready which means they can be used immediately with no additional effort required. A data product also includes the important meta data that helps to describe the data to provide business meaning. This is in contrast to how data was extracted in the past, which was simply as unpackaged, raw data that needed preparing before it could be used in the development of analytical applications. Just like currency, data products can grow in value. You achieve this by enriching data products using AI and machine learning and by feeding them back into the shared catalog for reuse and further enrichment.

A key aspect of a data product economy is that all data products should be discoverable and highly reusable.

SAP BDC provides certified, ready-to-use data products that are generated from all SAP applications and provide real-time data to analytical applications. SAP have done all the hard work to get data packed as a product ready to be used. Business users simply choose which SAP data product they would like to use from their SAP BDC catalog to build analytics.

SAP provide more than 100 different applications that cover all lines of business (LoB). The definition of a business entity such as a customer, might be provided by more than one SAP application. Unfortunately these definitions are sometimes not technically aligned so there might be two or more versions of customer. SAP BDC harmonizes the different definitions of the same business entity into one. This is referred to as the one domain model. As data flows from the source applications to SAP BDC, the data passes through a harmonization stage that ensures all versions of the same entity are unified. This means that the business user does not have to manage multiple versions of the same entity, and in fact does not need to know or care about which source application the data comes from.

To support SAP's business data fabric, SAP BDC is built on the principle of openness across the entire data ecosystem. SAP has partnered with well-known data platform vendors to ensure customers can combine their SAP BDC solution with partner solutions for bi-directional data sharing.

SAP recognizes that many customers already use external data platforms to manage their non-SAP data. Customers want to integrate their external data platforms with SAP BDC to form a complete picture of their business data. This is especially important for AI / ML projects that require a complete data set on which to build intelligent applications. SAP has partnered with major data platform providers for bi-directional data sharing.

Delta share is an open standard technology that has been implemented throughout SAP BDC. Delta share used a zero-copy approach to sharing live data between SAP BDC and any consumer application. This means that multiple copies of SAP BDC data are avoided when a single version of data is shared by all applications. Once the data is extracted from SAP source applications and stored in SAP BDC, the data is governed by SAP BDC and shared securely with any consumer. SAP BDC keeps track of the consumers who are sharing the data. Only those consumers that have been granted permission to the data can access it.

SAP customers have collected a huge amount of data from their SAP applications. They also collect data from third-party applications. They need to store data at scale whilst avoiding escalating costs.

SAP BDC use the SAP HANA Data Lake for data storage instead of traditional databases. The SAP BDC data storage is based on files that can handle unlimited data growth. Storage is de-coupled from compute which means organizations can store huge amounts of data without incurring the very high costs of data processing. In the past customers would need to choose a combined compute/store size for their data.

SAP BDC is built to serve multiple personas of the organization.

SAP BDC provides a common data foundation that is used by all types of personas.

For example:

• SAP BDC is used by the LoB consumer / business users who need to access real-time data to support decision-making.
• SAP BDC provides the data to the data scientists and data engineers for their machine learning projects.
• SAP BDC provides data to the application developers who are building advanced AI-powered applications and who need to connect with up-to-date and trusted data sources from SAP applications.
• Business Analysts use SAP BDC to help them solve business problems using real-time data from across the enterprise.
• Data modelers who build models require ready-to consume data from all sources across the enterprise.
• SAP BDC is used by the administrator to provision data products and intelligent applications and enable business users.

All personas use the same data products safe in the knowledge that their data sets are consistent. The different personas do not need to keep copies of the data as a single copy of the data is shared from the lake storage layer. If the data is enriched, a new data product is generated and is added to the lake storage layer for others to consume.