Rag And Retrieval easy
You plan to index 10 million chunks using 1,536-dimension fp32 embeddings. Roughly how much raw vector storage is that, and what is a sensible lever to reduce it?
- A About 6 GB; vectors are small, so storage is rarely a concern at this scale.
- B About 60 GB; a lower dimension or int8 quantization cuts it several-fold.
- C About 600 GB; only a larger chunk size can meaningfully reduce the footprint.
- D About 60 GB; storage scales with chunk length, so shorter chunks shrink each vector.
Answer: B. About 60 GB; a lower dimension or int8 quantization cuts it several-fold.
Each vector is 1,536 x 4 bytes, about 6 KB, so 10 million vectors take about **60 GB** before index overhead. Reducing dimension or quantizing to int8 cuts memory up to 4x at a small cost in recall. Option D has the right total but the wrong lever: vector size depends on dimension, not on chunk length, and shorter chunks actually create more vectors.
One system design question a day, with spaced repetition so the ones you miss come back.
Get the appRelated questions
- You must index 100M chunks using a 3,072-dimension embedding in fp32, served from RAM via HNSW. The raw vectors alone take about 1.2 TB before graph overhead. You need to cut memory several-fold while losing at most a point or two of recall@10. Which approach fits?
- A multi-tenant SaaS app stores all tenants in one vector index with a tenant_id field. Queries fetch the top 20 chunks by similarity and then drop chunks from other tenants. Small tenants often get 0-2 results. Why does this happen, and what is the fix?
- Runbooks are split into fixed 300-token chunks with no overlap. When a procedure step starts near the end of one chunk and finishes in the next, answers come back incomplete. What is the lowest-cost fix?
- Your offline eval shows recall@5 of 92% on a labeled set, so the gold passage is in the context for nearly every question. Yet human graders mark 35% of answers wrong, and the wrong answers often ignore or contradict the gold passage. Where should the team invest next?
- Your product requires every RAG answer to include clickable source citations that users can verify. How should those citations be produced?
- A runbook RAG system uses fixed 500-token chunks with no overlap. Users who ask for a multi-step recovery procedure get answers that stop at step 4 of 9, and retrieval logs show the top hit ends mid-procedure. What is the most targeted fix?