The new innovations for the database service should help developers build and scale AI apps. The features provide apps with accurate and relevant information.
Spanner is Google Cloud’s distributed SQL relational database service. It separates compute resources and data storage, making it easier to scale processing resources. Many of Google’s own solutions, including YouTube and the search engine, also use it. This is partly because the service supports large volumes of data.
Andi Gutmans, GM & VP of Engineering, Databases, says that building AI apps also requires more knowledge about the data and its interconnections. In addition, companies “also need to make it easy for their users to search for the right data — not only via keyword search, but also by implementing AI-enabled semantic search,” Gutmans said.
Graph and vector
That’s why Google Cloud is now releasing Spanner Graph, adding graph processing to the service. Spanner Graph supports the graph query language (GQL) standard and provides interoperability with SQL for querying structured and connected data in a single operation. AI apps and the underlying foundation models can be enhanced with these graphs using RAG. This ensures that relevant data is used and improves the final output of the model.
Other additions to Spanner include full-text search and vector search. This new vector function is based on the ScaNN algorithm. This allows indexing and searching vector embeddings, which is useful for building semantic search functions.