[
https://issues.apache.org/jira/browse/LUCENE-9237?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17043837#comment-17043837
]
David Smiley commented on LUCENE-9237:
--------------------------------------
Were you able to do a comparison while keeping the term dictionary memory usage
equal? This will take some repeated tweaking of the parameters that
UniformSplit provides and then examine the size of the term dict files (or some
similar approach). Annoying; i know. Without doing this, we allow any
postings format to cheat by using memory gratuitously over its competitor. An
analogy is doing tour de france competition and not checking who is on drugs
:-D. Or at least allowing an equal amount of drugs for the contestants -- LOL I
amuse myself. Also, check that the on-heap vs off-heap FST usage is equivalent
amongst the contestants, as this is easily toggled by any format.
> Faster TermsEnum intersect for UniformSplit
> -------------------------------------------
>
> Key: LUCENE-9237
> URL: https://issues.apache.org/jira/browse/LUCENE-9237
> Project: Lucene - Core
> Issue Type: Improvement
> Reporter: Bruno Roustant
> Assignee: Bruno Roustant
> Priority: Major
> Time Spent: 20m
> Remaining Estimate: 0h
>
> New version of TermsEnum intersect for UniformSplit. It is 75% more efficient
> than the previous version for FuzzyQuery.
> Compared to BlockTree IntersectTermsEnum:
> - It is still slower for FuzzyQuery (-37%) but it is faster than the
> previous version (which was -65%).
> - It is slightly slower for WildcardQuery (-5%).
> - It is slightly faster for PrefixQuery (+5%). Sometimes benchmarks show
> more improvement (I've seen up to +17% a fourth of the time).
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
