Vector Similarity Search in Spring with Redis Stack
summary
Claim: a Java Spring app performs production-grade vector similarity search without Python by using Redis Stack for vector indexes and Redis OM Spring for entity mapping. Tools: DJL with PyTorch-backed models (retinaface for detection, SBERT-like encoders) to produce byte-array embeddings, stored in Redis and indexed with HNSW/L2. The demo loads 11,637 embeddings, runs K=6 nearest neighbors, and returns plausible celebrity matches; searches are fast, embedding generation is the main latency.
AI-written summary · about
chapters (7)
- 0:00Introduction and background
- 1:28Why vector similarity search matters
- 17:50Generating vectors with Java
- 25:25What is Redis Stack
- 33:47Redis OM Spring for Spring Data
- 39:43Facial similarity search demo
- 56:23Performance and Q&A
related talks
55mVector Similarity Search in Spring with Redis Stack
01h 01mVector Similarity Search in Spring
29mUnleashing the Power of Vector Search and Semantic Caching in Java
50mA fun and absurd introduction to Vector Databases
46mA fun and absurd introduction to Vector Databases
47mA fun and absurd introduction to Vector Databases
14mSearch on Steroids
48mA fun and absurd introduction to Vector Databases