Vector Databases

Vector databases are the retrieval layer behind semantic search, RAG systems, memory, and embedding-powered applications.

This handbook explains how vector search works from the algorithmic level up through storage, ingestion, scaling, and metadata filtering.

Articles

• How Vector Search Works — Embeddings, Distance, and the ANN Problem
• HNSW and IVF-PQ — The Algorithms Behind Vector Search
• Vector Database Storage and Ingestion — Segments, WAL, and Compaction
• Scaling Vector Search — Distributed Architecture and Metadata Filtering