The new AI battleground: open lakehouse platforms, not closed stacks
Enterprise data modernization for AI now centers on open lakehouse platforms that can unify fragmented, legacy and cloud data estates into a single, vendor-neutral foundation for agentic AI data integration and real-time analytics infrastructure, allowing organizations to combine heterogeneous sources without lock-in, heavy ETL pipelines or constant data movement while preserving governance, performance and explainability across the business. SAP’s completed acquisition of Dremio, an open, high-performance data lakehouse platform, is a clear signal that the AI race is shifting from model hype to data architecture. Instead of defending closed ecosystems, large vendors are being forced to acknowledge what enterprise teams have known for years: AI fails when data lives in isolated silos, proprietary formats and fragile ETL flows. The new competitive edge lies in letting SAP and non-SAP data coexist, query-ready, on open standards.
SAP–Dremio: a bet that open beats lock-in for agentic AI
By folding Dremio into SAP Business Data Cloud, SAP is betting that an open lakehouse platform is the only realistic path to scalable enterprise AI. The combined stack promises agentic AI and analytics on SAP and non-SAP data in real time, with no data movement or conversion and better economics for enterprise analytics. This is not a minor feature update; it is an admission that the classic “pull everything into our warehouse” strategy is too slow and too expensive for AI agents that must act on live operational data. SAP plans to make Dremio’s Apache Iceberg-native architecture the foundation of Business Data Cloud, allowing data to be stored and managed without tying customers to proprietary technologies and positioning the platform as an enterprise lakehouse for real-time analytics and agentic AI across complex environments. In practice, that means less ETL, more federated access, and fewer excuses when AI projects stall because data is “not ready”.
Real-time analytics infrastructure that respects the data you already have
The most compelling part of SAP’s move is its focus on using existing data where it already resides instead of forcing wholesale migration. Dremio enables organisations to query data in place, removing the dependency on traditional ETL pipelines and extensive replication, while SAP HANA Cloud’s in-memory capabilities cover transactional performance. Together, they aim to deliver analytical and AI workloads in real time with no data movement or conversion necessary and better economics than fixed, over-provisioned infrastructure. This is the kind of real-time analytics infrastructure enterprises have wanted for years: federated access across diverse sources, governed through a universal, open data catalog built on Apache Polaris and the Apache Iceberg REST Catalog API, acting as discovery and semantic layer for Business Data Cloud. It is a pragmatic response to a painful truth: fragmented estates, proprietary formats and disconnected systems have been one of the biggest barriers to enterprise AI adoption.
DataXel on Google Cloud: modernization becomes a product, not a multi-year ordeal
While SAP attacks the lakehouse side of the problem, Incedo’s DataXel tackles the equally stubborn barrier of legacy warehouses and ETL environments. DataXel is an agentic AI-powered data modernization platform that helps enterprises replace legacy data warehouses and ETL with scalable, AI-ready data foundations faster and with less risk. Its arrival on Google Cloud Marketplace shows that modernization itself is turning into a consumable product, not a bespoke multi-year consulting exercise. Unlike traditional approaches dependent on manual conversion and validation, DataXel unifies discovery, business rule extraction, code conversion, validation, optimization and governance in one configurable platform, cutting manual effort while preserving business logic and data integrity. According to Incedo, organizations using DataXel can achieve up to 50% faster modernization timelines, significant cost reductions, and trusted, analytics-ready data foundations for AI. This aligns with a simple reality: AI-ready data can’t wait for five-year migration programs.
Enterprises are choosing flexibility and interoperability over single-vendor control
The common thread between SAP’s Dremio acquisition and DataXel’s marketplace launch is not AI itself, but what enterprises now demand from the underlying data platforms. Teams are prioritizing flexibility and interoperability over single-vendor control, focusing on open lakehouse platforms, open table formats like Apache Iceberg and agentic AI data integration that works across SAP, Google Cloud and a mix of legacy systems without proprietary lock-in. SAP explicitly reinforces its commitment to open standards and ongoing support for the open-source technologies that underpin modern data platforms, enabling SAP and non-SAP data to coexist within a shared open foundation. Incedo, in turn, makes modernization portable across Teradata, Informatica, Oracle and other environments, including migrations to BigQuery and PySpark. The message is clear: the winners in enterprise AI will not be those who lock customers into monolithic stacks, but those who give them reliable, open choices for where and how their data lives.
