Vector Databases
Index families, filtered search, freshness and deletes, sharding, and the operational reality of billion-scale ANN.
11concepts
99flashcards
80minutes of reading
- 01 Measuring ANN Recall Honestly Why recall requires the exact search an index exists to avoid, how recall@k and 1-recall@1 differ, and why throughput reported without a recall axis is unfalsifiable.
- 02 Operating a Vector Store in Production The operational facts that decide whether a vector search system works, why memory is the binding constraint, and the migration nobody plans for.