VectorDB

A database system optimised for storing high-dimensional embedding vectors and performing fast approximate nearest-neighbour search. A vector database is a data storage system designed specifically to store, index, and query vector embeddings. Unlike relational databases that search by exact value or keyword, vector databases find the most semantically similar items using distance metrics such as cosine similarity or dot product.

Vector databases are a core infrastructure component in RAG systems: document embeddings are stored during ingestion, and the database returns the most relevant chunks for each user query at inference time.

Vector Database is a foundational concept in Infrastructure. A database system optimised for storing high-dimensional embedding vectors and performing fast approximate nearest-neighbour search.