SAP Data Provisioning Insights - Part 1

Welcome to SAP Data Provisioning Insights blog series! In this series, We will cover various aspects of data provisioning, including the latest techniques, benefits, and challenges of data virtualization and replication methods. Data provisioning is a broad term that means getting data from one system to another.

There are many reasons why data provisioning is important, including:

• Extracting data from business applications and sending it to a central data warehouse.
• Providing real-time access to data for analysis.
• Distributing data from a central system to local systems.
• Combining data from several systems into one central system.
• Keeping different systems updated with the same information.
• Migrating data from an old system to a new one.

Application Controlled Data Provisioning:

When we say application-controlled data provisioning, we mean that a specific application takes charge of how data moves. These applications help connect to different data sources and targets and set rules for how the data is transferred. Some examples of these applications are SAP Data Services, SAP Landscape Transformation, and SAP Datasphere.

In this method, the application pulls data from one database and sends it to another. It handles everything, including how to extract the data, how it flows, and how it gets loaded.You can think of it like a post office that moves letters from one address to another, following specific guidelines.

Using a dedicated data provisioning application is especially helpful when you have multiple data sources that use different technologies or come from various providers.

Database Controlled Data Provisioning:

When we talk about database-controlled data provisioning, we mean that the data flow is managed using tools that are part of the database itself.

The simplest tool for data provisioning might be an export and import tool to transfer data between databases. However, some databases, like SAP HANA, offer more advanced tools that can handle complicated data scenarios, such as combining, validating, and enriching data.

Using the built-in tools of a database to manage data provisioning means you don't need separate applications for this, which makes things simpler.

Data Virtualization:

Data provisioning doesn’t always mean we have to move data to a target system. Instead, data can be accessed from a source system and provided to a target system when it's needed. After the target system has read the data, the connection is closed, and no data is transferred. This process is called data virtualization.

Data Replication:

When you need to keep data up-to-date between systems, you use data replication. This means copying data from one system to another, usually without making any changes. The goal is for the target system to get the latest data whenever it changes in the source system, so applications on the target system always have the most current information.

Data Transformation:

If you need to change the data after it’s been extracted but before it goes to a target system, this is called data transformation. Transformation is a broad term that can include simple tasks like applying filters to load only the data you need. It can also mean creating new values from the source data.

For eg, concatenating the first name and last name while adding the name to the target system.