{"ref":"content:ac-federated-machine-learning-with-sap-datasphere","page":"/content/ac-federated-machine-learning-with-sap-datasphere","json":"/content/ac-federated-machine-learning-with-sap-datasphere.json","generated":"2026-10-26","drafted_with_ai":true,"entity":{"id":"ac-federated-machine-learning-with-sap-datasphere","title":"Federated Machine Learning with SAP Datasphere","url":"https://architecture.learning.sap.com/docs/ref-arch/17c8c5","content_type":"reference-architecture","explains":["capability:predictive-models","artifact:btp-datasphere"],"summary":"SAP's design for training predictive models on an external machine learning platform against [SAP Datasphere](artifact:btp-datasphere) data that is not replicated to that platform. The open-source FedML Python library runs in the platform's notebook, connects to Datasphere over Python/SQLDBC and reads the analytic models in which Datasphere unifies SAP and non-SAP sources; a trained model can then be deployed to [SAP AI Core](artifact:btp-ai-core), or in several sub-pages to [SAP BTP, Kyma runtime](artifact:btp-kyma-runtime), which is the [predictive models](capability:predictive-models) path where training happens outside SAP. Sub-pages cover Amazon SageMaker, Azure Machine Learning, Google Vertex AI, [Databricks](external:databricks), NVIDIA RAPIDS on GPUs and IBM watsonx.","short":"Training models on external ML platforms against SAP Datasphere data","authors":["actor:sap-architecture-center"],"published":"2025-01-23","sources":[{"url":"https://raw.githubusercontent.com/SAP/architecture-center/main/docs/ref-arch/RA0003/readme.md","title":"RA0003 readme.md in SAP/architecture-center","checked":"2026-09-15","quote":"The SAP Federated Machine Learning Python library (FedML) applies the Data Federation architecture of SAP Datasphere for intelligently sourcing SAP and non-SAP data for Machine Learning experiments, run on any Machine Learning platform, thereby removing the need for replication or data movement."},{"url":"https://raw.githubusercontent.com/SAP/architecture-center/main/docs/ref-arch/RA0003/readme.md","title":"RA0003 readme.md in SAP/architecture-center","checked":"2026-09-15","quote":"The SAP and non-SAP data are unified in SAP Datasphere analytical models and exposed for consumption outside of SAP Datasphere."},{"url":"https://raw.githubusercontent.com/SAP/architecture-center/main/docs/ref-arch/RA0003/readme.md","title":"RA0003 readme.md in SAP/architecture-center","checked":"2026-09-15","quote":"The FedML Python Library is imported directly into the external machine learning platform. FedML connects to SAP Datasphere via secure Python/SQLDBC connectivity and helps federate the critical business data needed for training models in these external ML platforms."},{"url":"https://raw.githubusercontent.com/SAP/architecture-center/main/docs/ref-arch/RA0003/readme.md","title":"RA0003 readme.md in SAP/architecture-center","checked":"2026-09-15","quote":"Models trained in external Machine learning platforms are optionally deployed in SAP AI Core for inferencing via FedML's seamless deployment integration."},{"url":"https://raw.githubusercontent.com/SAP/architecture-center/main/docs/ref-arch/RA0003/fedml-azure/readme.md","title":"RA0003 fedml-azure/readme.md in SAP/architecture-center","checked":"2026-09-15","quote":"Models trained in Azure ML can also be optionally deployed in SAP BTP Kyma for inferencing via FedML-Azure's seamless deployment integration."},{"url":"https://raw.githubusercontent.com/SAP/architecture-center/main/docs/ref-arch/RA0003/fedml-nvidia/readme.md","title":"RA0003 fedml-nvidia/readme.md in SAP/architecture-center","checked":"2026-09-15","quote":"FedML's Connectivity Core component supports reading data from semantic models of SAP Datasphere directly into CUDA cuDF dataframes."},{"url":"https://raw.githubusercontent.com/SAP/architecture-center/main/docs/ref-arch/RA0003/fedml-watsonx/readme.md","title":"RA0003 fedml-watsonx/readme.md in SAP/architecture-center","checked":"2026-09-15","quote":"FedML (fedml-dsp) can be used in notebooks inside the IBM watsonx.ai and Waston Studio environments."}],"drafted_with_ai":true,"provenance":{"discovered_via":[{"source":"source:curator"}],"first_seen":"2026-09-15"},"published_on":"source:architecture-center"},"readings":{"liveness":[{"label":"Checked against the source","tier":"unknown","evidence":"nobody has checked it"},{"label":"Source reachable","tier":"good","evidence":"up as of 2026-10-26"},{"label":"Source updated","tier":"fair","evidence":"2025-01-23 by the publisher"}],"compatibility":[{"label":"Version","tier":"unknown","evidence":"no version declared"},{"label":"Moved since?","tier":"good","evidence":"no recorded change after 2025-01-23"},{"label":"Findings","tier":"good","evidence":"none flagged"}],"reach":[{"value":null,"unit":"views"},{"value":null,"unit":"comments"},{"value":null,"unit":"likes"}]},"signals":{"crawled_at":"2026-10-26","reachable":true,"source_updated":"2025-01-23"},"series":null,"edges":{"authors":[{"ref":"actor:sap-architecture-center","name":"SAP Architecture Center","json":"/actors/sap-architecture-center.json"}]}}