Vector Search easy
Your index of 1 billion vectors will not fit in RAM even across your whole cluster. Which technique most directly reduces memory, and at what cost?
- A Product quantization, which compresses each vector into short codes at some cost to recall.
- B Increasing HNSW graph connectivity, which spreads vectors more evenly at some cost to latency.
- C Switching from cosine to L2 distance, which stores vectors more compactly at some cost to accuracy.
- D Post-filtering results by metadata, which reduces the vectors kept in memory at some cost to speed.
Answer: A. Product quantization, which compresses each vector into short codes at some cost to recall.
**Product quantization** can shrink a 6 KB vector to tens of bytes, which lets billions of vectors fit in memory. The cost is approximate distances and lower recall, which can be partly recovered by re-ranking top candidates with full vectors. More HNSW connectivity increases memory rather than reducing it. If the data already fits in RAM, skip PQ and keep full recall.
One system design question a day, with spaced repetition so the ones you miss come back.
Get the appRelated questions
- You are building an IVF index on 20 million vectors. A teammate asks why the build needs a 'training' step before any vectors can be added. What is the reason?
- An HNSW index serves a marketplace where about 30% of listings are updated or removed each week. Each update is applied as a delete plus an insert, and deletes mark tombstones. Over a month, recall@10 drifts from 97% to 89% and p99 latency rises at the same ef_search. What is happening, and what should you do?
- Users delete documents in your primary database, but the deleted documents keep appearing in vector search results for hours. The vector index is rebuilt nightly. What is the best fix?
- Your embedding model outputs unnormalized vectors whose magnitudes vary widely. A teammate wants to switch the index from cosine similarity to dot product because it is cheaper to compute. What is the consequence?
- You are setting an SLO for a new ANN-backed recommendation search over 50 million vectors. A product manager asks for 100% recall@10 with single-digit millisecond latency. What is a realistic response?
- A B2B SaaS app serves 500 customers, and every query searches only one customer's documents. One large customer runs heavy batch queries that slow down search for everyone. Which sharding approach fits best?