Elastic integrates Elasticsearch with NVIDIA AI Factory design

Elastic has confirmed that its Elasticsearch technology now integrates with the NVIDIA Enterprise AI Factory validated design to support organisations developing AI systems on their own infrastructure.

The integration positions Elasticsearch as a recommended vector database for enterprises aiming to construct and implement on-premises AI factories, providing the foundation for real-time intelligence in AI agents and generative AI applications.

Elastic is developing a plugin utilising NVIDIA cuVS, a technology designed to expedite the construction and querying of vector search indexes. This is expected to deliver faster and more precise results for both real-time and generative AI workloads.

"We are obsessed with building the best vector database in the market," said Ken Exner, Chief Product Officer at Elastic. "NVIDIA Enterprise AI Factory validated designs enable Elastic customers to unlock faster, more relevant insights from their data."

The importance of vector search and vector databases in supporting AI applications continues to grow within the industry. Elasticsearch is already widely used for such applications across the sector, supported by an active open source community.

Elastic has invested in optimising the performance of its vector database. Previous efforts have included hardware-accelerated CPU SIMD instructions, innovations in vector data compression with Better Binary Quantization, and performance improvements in Filtered HNSW, all aimed at increasing the effectiveness of indexing and querying at scale.

Pat Lee, Vice President, Strategic Enterprise Partnerships at NVIDIA, commented on the role of vector technology in AI. "Vector databases are transforming enterprise AI by making it easier for companies to find and use information quickly," he said. "With Elasticsearch and the NVIDIA Enterprise AI Factory reference design, enterprises can unlock deeper insights and deliver more relevant, real-time information to AI agents and generative AI applications."

Elastic stated that further details regarding its ongoing collaboration with NVIDIA, and developments associated with vector search on GPU, are available via its official blog.