mccullocht commented on issue #14758: URL: https://github.com/apache/lucene/issues/14758#issuecomment-3656721300
> We don't spend time scoring within "dark" areas of the graph. My concern here is that graph administrivia (reading in vectors, binary search to find their offsets, etc.) is costing too much here. Maybe we can spend some space to make exploring the graph (I ain't talking about scoring...just reading in the neighbors and then iterating without scoring) cheaper. Reading the edges on a vertex is likely memory latency bound in the same way that vector scoring is, although it is less amenable to bulk access. There is still the option to visit multiple vertices at the same time as is suggested by the DiskANN paper (pop N vectors from the candidate queue, visit them all, etc) which may also help hide latency if the graph is actually on disk. -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
