lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Michael McCandless (JIRA)" <>
Subject [jira] [Commented] (LUCENE-6548) Optimize BlockTree's Terms.intersect a bit for "very finite" automata
Date Mon, 22 Jun 2015 16:51:00 GMT


Michael McCandless commented on LUCENE-6548:

I'll commit the last patch soon; I think there are other things we can do here, but we can
explore them later and this is already a good speedup, at lease in the "very finite" (arcs
that typically have only one label, not a range) automata intersect case.

> Optimize BlockTree's Terms.intersect a bit for "very finite" automata
> ---------------------------------------------------------------------
>                 Key: LUCENE-6548
>                 URL:
>             Project: Lucene - Core
>          Issue Type: Improvement
>            Reporter: Michael McCandless
>            Assignee: Michael McCandless
>             Fix For: 5.3, Trunk
>         Attachments: LUCENE-6548.patch, LUCENE-6548.patch, LUCENE-6548.patch
> I've been digging into why BlockTree's Terms.intersect impl is slower for a "very finite"
union-of-terms test over random IDs (LUCENE-3893) and I found a few performance improvements
that at least for that one use case gets a ~11% speedup.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message