a fundamental breakthrough in approximate nearest neighbor search for high-dimensional vector spaces. Introduced by Yury Malkov and Dmitry Yashunin in 2016. The algorithm achieves logarithmic search complexity through a sophisticated multi-layer graph structure that separates connections by characteristic distance scales, enabling efficient navigation from coarse to fine granularity during search operations. - Medium / Wikipedia

see also

• Scaling HNSWs - most of Redis is single-threaded, but antirez decided to use threads for the HNSW.
• How hierarchical navigable small world (HNSW) algorithms can improve search - everyone is connected by just a few people. That same principle powers hierarchical navigable small world (HNSW) algorithms, which link data points so queries can reach the right match in far fewer hops.