Skip to main content

FedML and IBM watsonx.ai / IBM Watson Studio integration

Last updated on by s-krishnamoorthy

This content has been archived

This content has been archived. It describes a retired or obsolete approach and is no longer maintained or supported by SAP. It remains available for reference only and should not be relied on as current guidance.

Joint Reference Architecture

This content was developed in collaboration with IBM.

FedML (fedml-dsp) can be used in notebooks inside the IBM watsonx.ai and Watson Studio environments.

FedML's Connectivity Core component supports reading data from semantic models of SAP Datasphere directly into the notebooks in IBM watsonx environments. FedML's IBM watsonx support helps data scientists accelerate machine learning workflows with IBM watsonx workflows, while providing instant access to SAP's critical business data without the need for ETL or additional overhead in processing ETL'd data.

Architecture​

image of solution diagram
Copy to clipboard
Solution Diagram Resources
You can download the Solution Diagram as a .drawio file for offline use. Alternatively, you may view and edit the Solution Diagram directly on draw.io.
Please note that any changes made online will need to be saved locally if you wish to keep them.
DownloadEdit
Resources

When to use​

  1. When models need to be trained inside the IBM watsonx.ai or IBM Watson Studio environments using SAP business data from SAP applications.
  2. When critical SAP data from various SAP applications (with semantics intact) is needed for such training jobs, FedML helps read the SAP data (via the SAP Datasphere semantic models) directly into the IBM Watson environment.
  3. Models trained in IBM Watson environment can be deployed in SAP AI Core for quick inferencing based on SAP business data.

Resources​