Hibernate Search 7: What's New and Practical Cases for Java Full-Text and Vector Search
summary
Hibernate Search 7.1 adds vector (embedding) fields and KNN/ANN searches integrated into full‑text indexes, demonstrated on Quarkus with Postgres and Elasticsearch using a Harry Potter dataset; vectors can be configured for dimension and distance metric and return different results than plain full‑text. It covers automatic denormalization from JPA entities to indexed documents, a transactional outbox that writes events to SQL to ensure eventual indexing, and a standalone POJO mapper plus Infinispan integration; Lucene backend also supports vector search and ANN is claimed to scale to large corpora.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 3:07Full-text search problems
- 7:21Hibernate Search demo
- 14:22Denormalization and indexing
- 17:56Hibernate Search 6 revolution
- 21:50Transactional Outbox
- 26:46Standalone POJO Mapper
- 27:49Infinispan and vector search
- 29:14Vector search principles
- 32:25Vector search demo
- 38:33Conclusion
related talks
51mHibernate Search talk
41mApache Solr 10: What's Coming up for Vector Search
44mVector Search Beyond the Hype: Real-World Implementation and Performance Insights
01h 09mVector Search: Ask Me Anything!
48mFrom Text to Meaning: pgvector Transforms PostgreSQL Search
25mFrom Full-Text to Vector Search: Modern Search Architectures for Java Applications
40mUnder the hood of vector search with JVector
46mNatural Language & Search, Improve your Search w/ Vector capabilities