kaivalnp commented on PR #14178: URL: https://github.com/apache/lucene/pull/14178#issuecomment-2710218340
Thanks @benwtrent! > While I think the performance numbers are cool, they indicate that this doesn't actually buy us that much The speedup we see above is just a pure HNSW v/s HNSW comparison, but IMO the main benefit of integrating Faiss is the ability to use different indexing and search strategies like vector transforms (eg PCA, OPQ, ITQ, etc), different indexes (eg IVF, PQ, HNSW+PQ, NSG, etc), refined searches (like a 2-step search), and hardware (eg GPUs, see https://github.com/facebookresearch/faiss/wiki/Faiss-on-the-GPU) out-of-the-box, along with any future advancements > as long as building sandbox doesn't require a new dependency +1, the PR does not add Faiss as a Lucene dependency -- and only requires shared libraries at runtime. We could create a GitHub action that builds the C_API and executes Faiss-specific tests in an automated way Assuming we can bring this PR to such a state where users need not build Faiss unless required (and it is still tested with each commit) -- do you think we can merge this codec? -- 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
