opensearch-project
[DOC] Lucene Inbuilt Byte Quantization
What do you want to do?
- [ ] Request a change to existing documentation
- [x] Add new documentation
- [ ] Report a technical problem with the documentation
- [ ] Other
Tell us about your request. Provide a summary of the request and all versions that are affected. Since OpenSearch version 2.9, users can ingest byte vectors using the Lucene Byte Vector feature. But, Cx still need to do the quantization on their end to convert floats into bytes as it only accepts the byte range [-128 to 127]. But, now lucene added support for inbuilt scalar quantization where it takes float vectors as input and quantizes them into byte vectors which removes the overhead for the Cx besides improving the performance and memory optimization.
What other resources are available? Provide links to related issues, POCs, steps for testing, etc. https://github.com/opensearch-project/k-NN/issues/1277
Hi @naveentatikonda , what is your ETA for producing the doc PR? To meet entrance criteria, the PR must be open and in review by 4/30. Thanks!
