IBM buys DataStax to boost its watsonx data platform for AI applications

IBM Corp. said today it’s buying the database company DataStax Inc. in an effort to turbocharge the capabilities of its growing watsonx portfolio of artificial intelligence development tools.

The company said it’s planning to integrate DataStax’s technology into watsonx’s generative AI products, where it will help manage and streamline access to the vast amounts of data they require. In addition, the deal will enhance its commitment to building open-source AI products.

The financial terms of the deal were not disclosed. It’s expected to close sometime during the second quarter, subject to regulatory approval.

DataStax’s flagship product is a cloud-based database-as-a-service offering called Astra DB, which is built on the popular open-source Apache Cassandra database management system that’s designed to handle vast amounts of information distributed across multiple locations and environments.

IBM stressed that DataStax will continue to work on the Apache Cassandra project, where it is one of the largest contributors, as well as the other open-source initiatives it’s involved in, such as Langflow, Apache Pulsar and OpenSearch.

The immediate plan will be to integrate Astra DB with its watsonx.data platform, which is a data lake for AI and analytics workloads. The integration will add vital vector search capabilities to watsonx.data, IBM said. In addition, it’s also interested in Langflow, which is a visual framework for building AI applications that can enhance its own AI development studio tools.

IBM Software Senior Vice President Dinesh Nirmal said in a blog post that businesses will struggle to realize the full potential of generative AI without the right infrastructure in place. That infrastructure consists primarily of open-source tools and technologies that empower developers and get a lid on unstructured data, he explained.

“DataStax possesses deep competency in this area and shares IBM’s relentless commitment to simplifying and scaling generative AI for the enterprise,” he added.

DataStax Chief Executive Chet Kapoor added in his own blog post that he has long said there is no AI without data. “This vision will now be propelled as DataStax joins IBM,” he said. “With our technology and IBM’s watsonx.data, we will be able to bring vector and AI search to the entire data estate and make IBM’s capabilities available to every developer.”

Constellation Research Inc. analyst Doug Henschen noted that IBM is getting a lot more for its money than just an enhanced AI data platform, as DataStax’s technology underpins the entire data environments of dozens of big enterprises. They include Netflix Inc., Overstock Inc., FedEx Corp., Capital One Financial Corp. and Intuit Inc. “The underlying platform is solid and geared to massive, global-scale deployments,” he pointed out.

That said, IBM may face a challenge in keeping some of those customers, for Henschen said some of them may instead turn to self-managing the open-source version of Apache Cassandra instead, or alternatively go with one of its rivals. “DataStax has faced increased competition in recent years, primarily from public cloud vendors, leading with AWS, which offers both DynamoDB and Amazon Keyspaces,” he explained.

Other tools within DataStax’s portfolio include the Astra Streaming service that’s used to create real-time data pipelines, plus the DataStax AI Platform for designing and developing AI applications. The company has also developed RAGStack, which helps simplify retrieval-augmented generation, a technique that enables large language models to access external data sources to improve their knowledge.

Last year, DataStax debuted a new enterprise-grade AI development platform that integrates Nvidia Corp.’s AI Enterprise software suite. It’s called the DataStax AI Platform, Built with Nvidia, and the company claims it’s one of the most comprehensive platforms for enterprise AI software development in the business.

Image: DataStax