Database
Amazon Aurora
Amazon Aurora is a MySQL- and PostgreSQL-compatible relational database built for managed availability and scale. Aurora PostgreSQL with pgvector can keep transactional metadata and vector retrieval near relational access controls.
Key points
- Storage replicates across multiple Availability Zones
- Aurora Serverless v2 adjusts database capacity within configured bounds
- PostgreSQL-compatible clusters can use supported vector extensions
When to use it
- Store tenant-scoped conversation state with relational constraints
- Join product entitlements to vector-search metadata for RAG
Exam tips
- Choose Aurora when transactions and SQL joins matter; OpenSearch is stronger for search-centric hybrid retrieval
- Index vectors and connection pools deliberately because serverless capacity does not remove query design