• Why JVector 3 Is The Most Advanced Embedded Vector Search Engine

  • Oct 13 2024
  • Length: 54 mins
  • Podcast

Why JVector 3 Is The Most Advanced Embedded Vector Search Engine

Listen for free

View show details

  • Summary

  • An airhacks.fm conversation with Jonathan Ellis (@spyced) about: discussion of JVector 3 features and improvements, compression techniques for vector indexes, binary quantization vs product quantization, anisotropic product quantization for improved accuracy, indexing Wikipedia example, Cassandra integration, SIMD acceleration with Fused ADC, optimization with Chronicle Map, E5 embedding models, comparison of CPU vs GPU for vector search, implementation details and low-level optimizations in Java, use of Java Panama API and foreign function interface, JVector's performance advantages, memory footprint reduction, integration with Cassandra and Astra DB, challenges of vector search at scale, trade-offs between RAM usage and CPU performance, Eventual Consistency in distributed vector search, comparison of different embedding models and their accuracy, importance of re-ranking in vector search, advantages of JVector over other vector search implementations

    Jonathan Ellis on twitter: @spyced
    Show More Show Less
Show More Show Less
activate_Holiday_promo_in_buybox_DT_T2

What listeners say about Why JVector 3 Is The Most Advanced Embedded Vector Search Engine

Average customer ratings

Reviews - Please select the tabs below to change the source of reviews.

Audible.co.uk reviews

Amazon reviews
No Reviews are Available