Analytics Legends The knowledge platform for SAP Analytics
Academy module

Lakehouse Patterns

Lakehouse Patterns: medallion architecture from SAP source through bronze, silver, and gold, feeding the Datasphere semantic layer and the Databricks Unity Catalog inside SAP Business Data Cloud — architecture diagram for Lakehouse Patterns, Analytics Legends Academy module M123

As of 2026-08-16

Edited by , sole editor and publication director — 27 years in the SAP ecosystem. How these pages are produced, and where automation is used.

Lakehouse Patterns settles the one architecture question every SAP Business Data Cloud (BDC) engagement hits in week one: does Databricks or Datasphere own governance, and where does SAP-specific logic — MANDT filtering, currency conversion, status decoding — actually live. Get the medallion boundary wrong (business logic in bronze instead of silver) and the client re-extracts from SAP every time a rule changes; get the BDC pattern wrong and Unity Catalog authorisation drifts from SAP's profit-centre model. The economics are concrete: Databricks on Delta runs roughly 4-6x cheaper than HANA Cloud in-memory for cold historical volumes above 5-50 TB, and architects who own the full stack — SAP extraction, Delta Live Tables, Unity Catalog, Datasphere semantic reconnection — bill at 900-1,200 EUR/day freelance in DACH and Benelux. This module gives the decision frame, not a product tour.

What you will learn

  • Explain the medallion (bronze-silver-gold) architecture and where it differs from a classic EDW star schema
  • Design a lakehouse integration for SAP sources using BDC + Databricks or Datasphere's lake layer
  • Choose between lakehouse and EDW approaches based on workload characteristics and governance constraints
  • Articulate the architectural trade-offs in client conversations and salary or rate negotiations

What a Lakehouse Actually Is (and Isn't)

The term "lakehouse" is overloaded. In the SAP context it means two distinct things depending on which SAP product you are standing in, and conflating them in a client conversation is an immediate credibility risk.

The Databricks/Delta Lake meaning -- open-format columnar storage (Delta, Iceberg, or Hudi) sitting on cheap object storage (ADLS Gen2, S3, GCS), with ACID transactions, schema enforcement, and a compute layer (Spark, Photon, DuckDB) that can run directly against those files without a proprietary warehouse format. A lakehouse in this sense replaces both the staging area and the data warehouse in a single physical tier, eliminating the ETL hop from lake to warehouse.

The SAP BDC meaning -- Business Data Cloud (GA 2025) packages Databricks as a first-class compute and storage tier alongside SAP Datasphere. Within BDC, the Databricks lakehouse is the open-format foundation; SAP Datasphere sits above it providing governed semantic layer, business content, and identity federation. You will encounter clients who have bought BDC but have not yet decided which workloads live in Databricks vs. Datasphere.

Prerequisites

  • Intermediate hands-on experience on SAP analytics projects
  • Review core concepts first: C035, C016, C087

Outcomes

  • Understand the core concepts behind lakehouse patterns
  • Apply Lakehouse in a typical SAP analytics engagement
  • Explain the core architecture and decision points for Lakehouse Patterns
  • Apply a repeatable implementation pattern in a 15-minute lab format

Full module available to members. The full module adds: the decision framework · the end-to-end scenario walkthrough · the KPI scorecard · the anti-patterns · the code blocks · the knowledge check · the diagrams.

Open in the app →

Next in this track

Guides that answer with this page

These guides cite this page as one of the sources their answer rests on.