rmuir commented on PR #11860: URL: https://github.com/apache/lucene/pull/11860#issuecomment-1301189426
ok, i understand the issue with reset, i think? But I am still unsure if reset is the correct solution: reset solution only works if the bpv is the same everywhere, which it happens to be the case now, but thats really strange? Should the whole graph really be encoded with the worst-case bpv? or would it be more efficient if it encoded each block independently like postings? Postings uses block size=128, and AFAIK previously used some lower-level FOR primitive, but nowadays reimplements the decode with hacks to trick the compiler's autovectorizer. But something like that might be the correct approach for fast decode and more space savings. Sorry, I don't know the answer myself because I don't know anything about the encoded data's distribution. -- 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
