Why Vector Storage

1. What is the main challenge associated with nearest-neighbor search in vector storage?

Finding exact matches using relational indices Identifying the nearest stored vectors among millions Sorting vectors based on their length Reducing the size of the data stored in vectors

2. Which distance measure is commonly used for text embeddings to determine similarity?

Euclidean distance Manhattan distance Hamming distance Cosine distance

3. Why are traditional B-tree indices ineffective for searching in vector spaces?

They require exact matches to function They can only store single-dimensional data They cannot handle unordered multi-dimensional data They do not support fast search algorithms

4. What architectural choice does LyraLearn use to store its vector embeddings?

A separate vector database service An in-memory data structure A SQL Server 2025 column alongside relational data A NoSQL document store

Continue: Vector Indexes → Review this lesson Retake quiz