asfimport
asfimport
Adrien Grand (@jpountz) ([migrated from JIRA](https://issues.apache.org/jira/browse/LUCENE-8542?focusedCommentId=16790497&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16790497)) I think this goes against the use-case that Lucene is designed to solve indeed, which is to compute the top-k matches of a query...
Christoph Kaser ([migrated from JIRA](https://issues.apache.org/jira/browse/LUCENE-8542?focusedCommentId=16790560&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16790560)) That's too bad, given that this is only a minor change to an experimental API (and does not cause extra work in the reasonable use...
Michael McCandless (@mikemccand) ([migrated from JIRA](https://issues.apache.org/jira/browse/LUCENE-8542?focusedCommentId=16790576&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16790576)) I think the core API change is quite minor and reasonable – letting the `Collector.newCollector` know which segments (slice) it will collect? E.g. we...
Adrien Grand (@jpountz) ([migrated from JIRA](https://issues.apache.org/jira/browse/LUCENE-8542?focusedCommentId=16790607&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16790607)) I don't think this change really helps as the number of documents in a slice is a pretty bad upper bound of the total...
Christoph Kaser ([migrated from JIRA](https://issues.apache.org/jira/browse/LUCENE-8542?focusedCommentId=16790611&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16790611)) While it's true the slice size is a bad upper bound, the change does help: As you can see in the [table in my comment](https://issues.apache.org/jira/browse/LUCENE-8542?focusedCommentId=16704391&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16704391),...
Michael McCandless (@mikemccand) ([migrated from JIRA](https://issues.apache.org/jira/browse/LUCENE-8542?focusedCommentId=16790677&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16790677)) Maybe we should try swapping in the JDK's `PriorityQueue` and measure if this really hurts search throughput?
Adrien Grand (@jpountz) ([migrated from JIRA](https://issues.apache.org/jira/browse/LUCENE-8542?focusedCommentId=16790882&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16790882)) Right I get how it can help with small slices, but at the same time I'm seeing small slices as something that should be...
Lucene/Solr QA ([migrated from JIRA](https://issues.apache.org/jira/browse/LUCENE-8542?focusedCommentId=16791357&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16791357)) |✔ **+1 overall**| |-| |Vote|Subsystem|Runtime|Comment| |-|-|-|-| |||| Prechecks | |+1| test4tests | 0m 0s| The patch appears to include 1 new or modified test files....
Christoph Kaser ([migrated from JIRA](https://issues.apache.org/jira/browse/LUCENE-8542?focusedCommentId=16791424&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16791424)) > Right I get how it can help with small slices, but at the same time I'm seeing small slices as something that should be...
Adrien Grand (@jpountz) ([migrated from JIRA](https://issues.apache.org/jira/browse/LUCENE-8542?focusedCommentId=16791471&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16791471)) > the default implementation of slices() may not be optimal. Let's change it?