Re: [DISCUSS] Add JVector as a dependency for CEP-30

[J. D. Jordan]

+1 for jvector rather than forked lucene classes.  On Sep 20, 2023, at 5:14 PM, German Eichberger via dev <dev@cassandra.apache.org> wrote:  +1 I am biased because DiskANN is from Microsoft Research but it's  a good library/algorithm From: Mike Adamson <madam...@datastax.com> Sent: Wednesday, September 20, 2023 8:58 AM To: dev <dev@cassandra.apache.org> Subject: [EXTERNAL] [DISCUSS] Add JVector as a dependency for CEP-30   You don't often get email from madam...@datastax.com. Learn why this is important The original patch for CEP-30 brought several modified Lucene classes in-tree to implement the concurrent HNSW graph used by the vector index. These classes are now being replaced with the io.github.jbellis.jvector library, which contains an improved diskANN implementation for the on-disk graph format.  The repo for this library is here: https://github.com/jbellis/jvector. The library does not replace any code used by SAI or other parts of the codebase and is used solely by the vector index. I would welcome any feedback on this change. -- Mike Adamson Engineering +1 650 389 6000 | datastax.com Find DataStax Online: