Database / Weaviate Vector database Interview questions
What is the HNSW index in Weaviate?
HNSW (Hierarchical Navigable Small World) is Weaviate's primary, in-memory graph-based approximate nearest neighbor index, and historically its only index type, still the default and most commonly used choice for collections of meaningful size.
vector_index_config=Configure.VectorIndex.hnsw( ef_construction=128, max_connections=32 )
Because HNSW keeps its graph structure in memory, it delivers fast, high-recall search, but its memory footprint scales with dataset size, which is why Weaviate offers several compression techniques (PQ, BQ, SQ, RQ) specifically for HNSW indexes, letting a collection reduce memory usage substantially in exchange for a small, generally well-controlled amount of recall loss.
HNSW in Weaviate is:
Compression techniques like PQ, BQ, SQ, and RQ exist for HNSW mainly to:
More Related questions...
