Back to Devoured - May 07, 2026
DEVOURED
SAP to acquire data lakehouse vendor Dremio

SAP to acquire data lakehouse vendor Dremio

Data aienterpriseacquisition
SAP is acquiring Dremio to capture customer data that manufacturing and regulated industries refuse to move to cloud, fixing a strategic error by adopting Apache Iceberg after years of pushing SAP-only data hosting.
What: SAP announced plans to acquire data lakehouse vendor Dremio for an undisclosed price to make SAP Business Data Cloud Apache Iceberg-native. SAP CTO Philipp Herzig said this addresses LLM limitations with structured data and predictive analytics. Unlike SAP partners Snowflake and Databricks, which require data migration and reformatting, Dremio provides federated access to data in place. Analysts including Aman Mahapatra (Tribeca Softtech CSO) and Jason Andersen (Moor Insights & Strategy) noted this lets SAP reach data in manufacturing and regulated industries that refuse cloud migration.
Why it matters: Signals SAP's strategic pivot from forcing data migration to federated access that meets enterprises where they are. Multiple analysts said SAP is correcting a years-old error by finally adopting Iceberg, and by H1 2027, SAP will likely deprioritize Snowflake and Databricks integrations despite current partnerships, as the defensible value in enterprise AI migrates from compute to the semantic layer.
Takeaway: If you run SAP-heavy analytics on Snowflake or Databricks, expect SAP to steer new AI workloads toward Business Data Cloud by H1 2027. Negotiate contracts now while you have leverage.
Deep dive
  • SAP announced plans to acquire Dremio, a data lakehouse vendor, for an undisclosed price to make SAP Business Data Cloud Apache Iceberg-native with federated access to SAP and non-SAP data
  • Key differentiator: Dremio accesses data in place within enterprise on-premises systems without requiring migration or reformatting, unlike Snowflake and Databricks which require moving data to their platforms first
  • SAP CTO Philipp Herzig said this addresses LLM limitations with structured data and predictive analytics, areas where LLMs struggle compared to unstructured text analysis
  • Critical for highly-regulated enterprises and manufacturing firms that refuse to move data to cloud, giving SAP access to pockets of customer data it couldn't reach with its cloud-first strategy
  • Implementation speed in days versus weeks or months compared to Snowflake and Databricks, though potentially at the expense of data processing performance
  • Multiple analysts characterized this as SAP correcting a strategic error by not developing Apache Iceberg capabilities years earlier, instead pushing customers toward SAP-only data hosting
  • Snowflake and Databricks are existing SAP partners, making the acquisition politically complex, but SAP maintains neutrality by positioning Dremio as a federated layer above the warehouse level
  • Analyst Aman Mahapatra predicts that by H1 2027, SAP will steer net-new AI workloads toward Business Data Cloud regardless of partnership press releases, making warehouse vendors less strategic
  • Analyst Jason Andersen noted this is face-saving for SAP to access data without reversing its years-long position of encouraging enterprises to host all data within SAP systems
  • Defensible value in enterprise AI is migrating from compute and storage (commoditizing) to semantic layer, catalog, lineage graph, and business context, where Dremio gives SAP ownership
  • Most large SAP customer estates are brownfield landscapes with distributed data across SAP, non-SAP, legacy, departmental, regional, acquired, and partner systems, making migration-first approaches impractical
Decoder
  • Apache Iceberg: Open-source table format for large-scale analytical datasets that acts as a standardized bridge between raw data files and analytical tools, enabling interoperability across data platforms.
  • Data lakehouse: Architecture combining data lake storage (raw, unstructured data) with data warehouse capabilities (structured, query-optimized analytics).
  • Semantic layer: Abstraction layer defining business logic, metrics, and relationships between data entities so AI systems can interpret data meaningfully without understanding raw technical schemas.
Original article

SAP's Dremio acquisition is a pragmatic bet on AI-ready enterprise data, using Iceberg-native federated access to unify SAP and non-SAP data without major migration.

Explore all 73 articles in today's digest

Fresh dev news, AI-digested daily. Never miss a trend at devoured.fyi