PRODUCTION SCENARIO
A university archive is building a knowledge base over 9 million scanned pages. Researchers query it a few hundred times a day. Storage cost dominates the budget, and sub-second query latency is acceptable.
Which vector store will meet these requirements at the lowest cost?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: Amazon S3 Vectors
Amazon S3 Vectors is documented as best suited to infrequent query workloads and helps reduce the cost of retrieval augmented generation and semantic search, which matches a few hundred queries a day. Amazon OpenSearch Serverless and Amazon Aurora are supported stores but bill for query capacity that stays provisioned, and Amazon Neptune Analytics backs graph retrieval the archive does not need.