Much like the emergence of cloud computing, agentic artificial intelligence is causing many data platform software providers to fundamentally rethink value propositions. For example, the ability to accelerate data workloads with custom hardware was an important differentiator in an era when almost all data processing was performed using on-premises server infrastructure but quickly became a niche proposition in a world dominated by low-cost, general-purpose cloud compute services. Similarly, the need for a high-performance analytics engine for business intelligence acceleration is quickly becoming marginalized as conversational and agentic AI interfaces become ubiquitous. Providers previously defined by the ability to accelerate traditional analytics workloads, such as Exasol, are adapting products and positioning accordingly.
Exasol was founded in Germany in 2000 to develop an in-memory columnar database software specifically for analytics. Based on a shared-nothing architecture, Exasol Database was designed to enable enterprises to distribute queries across various nodes in a cluster using optimized parallel algorithms to process data locally. While Exasol can be used as a standalone data warehouse, the company has in recent years more often positioned it as an in-memory complement to an existing data warehouse—deployed on-premises or in the cloud. Exasol’s Virtual Schemas abstraction layer is used to access and query external data sources, enabling it to accelerate data lakehouse environments or unify data from multiple analytic data platforms without data movement.
The company was named Exemplary in ISG’s 2026 Buyers Guide for Data Platforms Emerging Providers, which included providers with annual revenue of between $25 million and $75 million. Exasol reported total revenue of €41.8 million in 2025, up 5.6% on the previous year. 2025 was a year of realignment for the company, as it focused its resources and sales efforts on key industries, including finance, healthcare, telecommunications, utilities and the public sector. Annual recurring revenue from these key industries rose by 10.1% in 2025. As I recently noted, the company also entered a potentially significant partnership with MariaDB in 2025, with the open-source database provider offering the Exasol Analytics Engine to the MariaDB Enterprise Platform as MariaDB Exa.
Exasol Database was designed to deliver high-level query performance. Traditionally, that meant high levels of concurrency and query performance for business intelligence dashboards and reports, as well as in-database predictive analytics via the training and deployment of machine learning models written in languages such as R, Python and Java. Today that same functionality is well-placed to accelerate agents and agentic applications. Exasol has added an open-source MCP server to provide connectivity to models, tools and agents via Model Context Protocol, enabling natural language querying of data and metadata discovery. The company has also added the Exasol Text AI extension to enable inference, keyword extraction and entity recognition from unstructured data using SQL-native user-defined functions, as well as the Exasol Semantic Layer to support centralized business logic and semantic definitions. As I recently explained, access to semantic models is required to provide a foundation of business context that can be understood and acted upon by autonomous agents.
In late 2025, Exasol announced the availability of Exasol Personal, a free but fully functional version of the product that can be deployed locally or on a customer’s cloud
account. While there are no data or cluster size limitations, Exasol Personal is limited to 20 connections and is intended for use by a single user. The full Exasol Enterprise product offers unlimited connections, an additional administrative user interface and is positioned to support sovereign AI and data strategies through a combination of on-premises, cloud and hybrid deployment options. I assert that through 2028, one-half of enterprises will prioritize data platform providers that offer a choice between self-managed or managed deployment options to support a sovereign AI and data strategy.
Exasol has also expanded its support for deployment as a complement to data lakehouse environments with the launch of Lakehouse Turbo, which is designed to accelerate Databricks analytics by providing sub-second analytics on queries in the Databricks Data Intelligence Platform without the need for extraction and transformation pipelines or data movement. The product is designed to connect to Databricks Unity Catalog and scan available metadata before creating a cache of selected tables required to support high-performance analytics use cases.
Exasol is well-known as a provider of high-performance data processing and analytics in Europe, especially in its native Germany. It currently has a limited profile in North America and will need to expand to break out of the emerging provider category. Positioning the product as a complement to data lakehouse environments should help, as should narrowing its focus to key industries that concentrate on high-performance analytics and AI. I recommend that enterprises looking to improve data stack performance and enable real-time agents consider Exasol in evaluations.
Regards,
Matt Aslett
Fill out the form to continue reading.