[
https://issues.apache.org/jira/browse/LUCENE-3328?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13068703#comment-13068703
]
Michael McCandless commented on LUCENE-3328:
--------------------------------------------
Results from latest patch:
||Task||QPS base||StdDev base||QPS bqterms||StdDev bqterms||Pct diff||||
|Fuzzy2|34.74|0.40|34.58|0.49|{color:red}2%{color}-{color:green}2%{color}|
|Fuzzy1|23.85|0.25|23.78|0.31|{color:red}2%{color}-{color:green}2%{color}|
|Respell|31.26|0.40|31.24|1.08|{color:red}4%{color}-{color:green}4%{color}|
|PKLookup|137.63|2.79|138.19|6.90|{color:red}6%{color}-{color:green}7%{color}|
|OrHighHigh|18.91|0.29|19.04|0.49|{color:red}3%{color}-{color:green}4%{color}|
|Wildcard|50.35|1.12|50.82|1.32|{color:red}3%{color}-{color:green}5%{color}|
|IntNRQ|9.14|0.38|9.24|0.42|{color:red}7%{color}-{color:green}10%{color}|
|OrHighMed|10.38|0.36|10.55|0.39|{color:red}5%{color}-{color:green}9%{color}|
|Prefix3|15.22|0.34|15.51|0.30|{color:red}2%{color}-{color:green}6%{color}|
|TermBGroup1M1P|57.45|2.11|58.81|1.75|{color:red}4%{color}-{color:green}9%{color}|
|SpanNear|4.28|0.28|4.40|0.08|{color:red}5%{color}-{color:green}11%{color}|
|TermBGroup1M|46.36|1.38|47.93|1.88|{color:red}3%{color}-{color:green}10%{color}|
|TermGroup1M|47.63|0.93|49.31|1.83|{color:red}2%{color}-{color:green}9%{color}|
|Term|137.86|6.74|142.82|6.55|{color:red}5%{color}-{color:green}13%{color}|
|SloppyPhrase|19.72|0.87|20.62|0.01|{color:green}0%{color}-{color:green}9%{color}|
|Phrase|3.28|0.19|3.50|0.11|{color:red}2%{color}-{color:green}16%{color}|
|AndHighMed|87.52|2.55|100.80|4.80|{color:green}6%{color}-{color:green}24%{color}|
|AndHighHigh|14.92|0.55|17.33|0.91|{color:green}6%{color}-{color:green}26%{color}|
Looks great!
> Specialize BooleanQuery if all clauses are TermQueries
> ------------------------------------------------------
>
> Key: LUCENE-3328
> URL: https://issues.apache.org/jira/browse/LUCENE-3328
> Project: Lucene - Java
> Issue Type: Improvement
> Components: core/search
> Affects Versions: 3.4, 4.0
> Reporter: Simon Willnauer
> Fix For: 4.0
>
> Attachments: LUCENE-3328.patch, LUCENE-3328.patch, LUCENE-3328.patch
>
>
> During work on LUCENE-3319 I ran into issues with BooleanQuery compared to
> PhraseQuery in the exact case. If I disable scoring on PhraseQuery and bypass
> the position matching, essentially doing a conjunction match,
> ExactPhraseScorer beats plain boolean scorer by 40% which is a sizeable gain.
> I converted a ConjunctionScorer to use DocsEnum directly but still didn't get
> all the 40% from PhraseQuery. Yet, it turned out with further optimizations
> this gets very close to PhraseQuery. The biggest gain here came from
> converting the hand crafted loop in ConjunctionScorer#doNext to a for loop
> which seems to be less confusing to hotspot. In this particular case I think
> code specialization makes lots of sense since BQ with TQ is by far one of the
> most common queries.
> I will upload a patch shortly
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
