jbellis commented on issue #12615: URL: https://github.com/apache/lucene/issues/12615#issuecomment-1772711571
> DiskANN is known to be slower at indexing than HNSW I don't remember the numbers here, maybe 10% slower? It wasn't material enough to make me worry about it. (This is with using an HNSW-style incremental build, not the two-pass build described in the paper.) > and the blog post does not compare single threaded index times with Lucene. It was about 30% worse several months ago with my concurrent hnsw patch, should be significantly improved now since JVector (1) doesn't need the CompletionTracker to keep the layers consistent, b/c it's single layer, (2) added a DenseIntMap concurrent collection to replace ConcurrentHashMap for the nodes. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org