Skip to main content

Integration with Palantir

Last updated on by s-krishnamoorthy

Overview​

Palantir Foundry is an enterprise data and AI platform used by large organizations to integrate operational data, build AI/ML pipelines, and operationalize decisions.

For SAP customers using Palantir Foundry, SAP Data Accelerator is the SAP-endorsed connectivity service providing a preferred, high-throughput, governed path for ingesting SAP data into Palantir Foundry.

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

Architecture Details​

Palantir Foundry's SAP integration stack includes these deployment components.

  • SAP Data Accelerator: The core SAP service that provides the secure data path between the customer's SAP landscape and Palantir Foundry.
  • Backend SAP Systems: Data sources — on-premises and SAP S/4HANA Cloud Private Edition.
  • Foundry Connector (Palantir Foundry Connector 2.0 for SAP Applications): Installed in the SAP backend system, this add-on registers extraction logic and manages delta queue operations, exposing data to SAP Data Accelerator in a structured format.
  • SAP Cloud Connector: Establishes a secure, outbound-only tunnel from on-premises or SAP S/4HANA Cloud Private Edition systems to SAP Data Accelerator — requiring no inbound firewall openings on the SAP side.
  • Palantir Foundry: Partner platform consuming SAP data via the SAP Data Accelerator Hub.

Existing users of Palantir's native SAP connectors can transition seamlessly to the SAP Data Accelerator architecture while retaining their existing Foundry Connector installation — no re-installation or re-configuration of the add-on is required.

Data Ingestion into Foundry​

Once the SAP Data Accelerator connector is configured in Foundry's connection management:

  1. Initial load: The full dataset is pulled from SAP via SAP Data Accelerator and landed as Foundry datasets with automatic schema detection.

  2. Incremental sync: Delta records from SAP Data Accelerator are applied to Foundry datasets, maintaining a current, deduplicated view of SAP data without re-extracting the full table.

  3. Schema registration: SAP objects surfaced through SAP Data Accelerator are registered in Foundry's data catalog with lineage tracking — users can trace every dataset back to its SAP source table or CDS view.

  4. Ontology mapping: Foundry's Ontology layer can be used to model SAP objects (e.g., SalesOrder, Material, Vendor) as typed Object Types, enabling cross-source joins and semantic search across SAP and non-SAP data.

Writeback from Foundry to SAP​

Foundry's writeback pipeline sends data through SAP Data Accelerator to the Foundry Connector, which handles the write to the SAP backend system. This enables user-attributed writebacks — changes committed to SAP are recorded against the Foundry user who triggered the action, not a shared service account. This is critical for audit compliance in regulated industries (financial services, pharma, public sector).

Typical writeback scenarios from Palantir Foundry to SAP include:

  • AI-optimized procurement recommendations → SAP purchase requisitions
  • Demand planning outputs → SAP planned independent requirements (PIRs)
  • Dynamic pricing decisions → SAP condition records
  • Workforce scheduling results → SAP HR time entries

Security and Governance in the Palantir Context​

  • Access control: Foundry’s fine-grained dataset access policies layer on top of SAP Data Accelerator’s SAP-side authorization. A user in Foundry can only query SAP-derived data if both Foundry and SAP Data Accelerator authorize the access.
  • Data lineage: All SAP data ingested via SAP Data Accelerator carries lineage metadata through Foundry pipelines — compliance teams can audit which SAP tables fed which AI model or report.
  • Encryption: Data is encrypted in transit (TLS) between SAP Cloud Connector, SAP Data Accelerator, and Foundry. Foundry applies encryption at rest within its managed storage.
  • User-attributed writeback: Writeback operations require explicit authorization — changes written to SAP are attributed to the individual user identity, not a shared service account, supporting audit compliance in regulated environments.

This reference architecture document is intended for enterprise architects and integration specialists evaluating or implementing SAP Data Accelerator as a governed data-sharing layer between SAP and partner platforms.