Advancements in SingleStore Database for AI Workloads

SingleStore has released an update to its database platform, SingleStore Pro Max, that enhances functionality for generative AI workloads, as well as transactional and analytics data workloads. The update includes new indexed vector search capabilities to support gen AI applications and retrieval augmented generation (RAG) use cases.

SingleStore has been offering vector capabilities in its database since 2017, when it was known as MemSQL. The company rebranded as SingleStore in 2020, becoming a unified platform that combines Online Analytical Processing (OLAP) and Online Transaction Processing (OLTP).

Importance of Vector Database Capabilities

With the rise of gen AI workloads, the need for vector database capabilities has also increased. Existing database vendors, including DataStax, Neo4j, MongoDB, PostgreSQL, and Oracle, have added vector support to their platforms. However, SingleStore CEO Raj Verma believes that solely relying on a purpose-built vector-only database is not the best approach.

“We provide you with a gen AI stack including vectors that allows you to build and model gen AI applications,” says Verma. “What we believe is that a vector-only database is a feature set and not a database that is going to be around in probably two maximum three years, because it adds a further layer of complexity in your AI stack and what you want to have an effective gen AI stack is to take complexity out, not add further complexity.”

According to Verma, SingleStore is focused on simplifying the data landscape for organizations. They aim to provide a vector database option as part of a larger converged data estate that includes other data types, enabling simplicity and speed for gen AI applications.

SingleStore’s Pro Max release brings advanced support for vector search across structured and unstructured data. The update includes faster and more accurate algorithms such as product quantization (PQ), Hierarchical Navigable Small World (HNSW), and Approximate Nearest Neighbor (ANN) vector indexing algorithms. These enhancements enable organizations to leverage their data effectively for search and gen AI applications.

While having a database dedicated to supporting vectors can help organizations quickly enter the gen AI space, Verma emphasizes that it doesn’t address the complexity of the broader data landscape. He believes that only through data consolidation and simplicity can organizations truly succeed in their gen AI endeavors.

It is common for organizations to have data stored in multiple databases. SingleStore’s Pro Max includes enhanced change data capture capabilities, allowing integration of data from MySQL and MongoDB databases, as well as Apache Iceberg-based data lakes, into a single database. The support for Apache Iceberg, an open-source data lake table format, facilitates easier integration with leading vendors like IBM and Snowflake.

“CDC capability allows our customers to have the ability to have the data from various sources brought into SingleStore which is extremely important for the entire retrieval, augment augmented generation workflow,” said Verma.

Leave a Reply

Your email address will not be published. Required fields are marked *

Related Posts