[
https://issues.apache.org/jira/browse/SOLR-9981?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16063489#comment-16063489
]
Steve Rowe commented on SOLR-9981:
----------------------------------
bq. However we are replacing this whole component with the second generation,
and this code will no longer be used. So fixing it isn't really necessary.
Maybe revert the commits on this issue then, given that it causes reproducible
failures?
> Multiple analytics fixes/performance improvements
> -------------------------------------------------
>
> Key: SOLR-9981
> URL: https://issues.apache.org/jira/browse/SOLR-9981
> Project: Solr
> Issue Type: Improvement
> Security Level: Public(Default Security Level. Issues are Public)
> Reporter: Houston Putman
> Assignee: Dennis Gove
> Priority: Minor
> Labels: patch
> Fix For: master (7.0)
>
> Attachments: SOLR-9983.patch
>
>
> Included are the following improvements/fixes:
> * Improving the unit test case.
> * Performance fix that stops the reading of ALL lucene segments over and
> again for each stats collector.
> ** The AtomicReaderContext that refers to the "current " segment is reused.
> ** This fix shows an improvement of about 25% in query time for a dataset of
> ~10M (=9.8M) records.
> ** Given the nature of the fix, the improvement should get better as the
> dataset increases.
> * Fix for the NPE during comparison
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
