Fabric Mirroring: How to efficiently integrate and utilise SAP data

SAP data is invaluable to businesses. It forms the basis for steering the business with precision and continuously optimising it at both the operational and strategic levels. Today, artificial intelligence and real-time analytics are opening up entirely new possibilities.

However, to make the most of this, the data must still be centrally provided and consolidated – regardless of whether it resides in S/4HANA, SAP BW(N) or ECC. So does setting up a modern data and AI solution necessarily involve a time-consuming and costly integration project? Not necessarily. In fact, current technologies also offer simpler ways to use SAP data in conjunction with other sources for a wide variety of use cases.

The best example is Microsoft Fabric. This innovative all-in-one platform is designed to be straightforward to use. With its ‘mirroring’ feature, the tool now offers a function that, when combined with SAP Datasphere, promises highly efficient integration of SAP data.

The benefits of Fabric Mirroring

Fabric’s mirroring function enables data from various source systems to be automatically mirrored or replicated in near real time to the central data store – the OneLake.
This approach offers several advantages over traditional ETL integration:

• no complex process implementation
• data provision without delays
• no additional infrastructure required
• no specialist knowledge required

In addition, the source data is replicated in an open format that can be used directly for analytics and AI workloads

Fabric Mirroring via SAP Datasphere

To integrate SAP sources into enterprise-wide data analytics using Fabric Mirroring, only one additional component is required: SAP Datasphere as the data management system. The technical process is kept very simple.

It consists of two steps:

1. Data replication via SAP Datasphere
   SAP Datasphere establishes a connection to the relevant SAP source system, extracts the relevant tables or views, and writes the data — initially and incrementally, or in a step-by-step manner — to an ADLS Gen2 storage container.
2. Setting up the mirrored SAP database in Fabric
   A shortcut is created in OneLake that points to the ADLS Gen2 container. The Mirrored SAP service is then used. Once the folder containing the replicated SAP data has been selected, Fabric creates a read-only, continuously synchronised copy of the SAP data as a “mirrored database”.

Finally, the SAP data is immediately available via the OneLake SQL endpoint. It can be linked to other sources and used for every conceivable application scenario, from BI and reporting right through to data science.

Conclusion

OneLake acts as a unified, scalable database with centralised management and data governance that can be used collaboratively by both IT and business departments. This allows SAP data to be quickly and easily linked with other data sources and analysed together. Traditional T-SQL queries, views and functions are created via the SQL endpoint.

This creates the conditions needed to fully realise the strategic value of SAP data – for informed decision-making, automated processes and data-driven innovation. At the same time, costs within Fabric remain manageable. There is no charge for the required compute resources. Mirroring storage is also free up to a certain limit. More detailed information on this can be found in the Fabric price overview.

However, there are still costs associated with the SAP Datasphere licence and the provision of the relevant data.

The mirroring function is currently still in the preview stage. Its use in production environments should therefore be carefully considered in light of governance and compliance requirements.

As a Fabric Featured Partner, ORAYLIS provides ongoing support to businesses in the expert implementation of modern Fabric solutions. If required, we can assist you every step of the way, from strategic planning through to day-to-day operations. Please get in touch if you too would like to fully leverage the business value of your data with Fabric.