[jira] [Commented] (SOLR-5478) Speed-up distributed search with high rows param or deep paging by transforming docId's to uniqueKey via memory docValues

["Manuel Lenormand (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SOLR-5478?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13827998#comment-13827998 ] 

Manuel Lenormand commented on SOLR-5478:
----------------------------------------

Following this mailing list thread - http://lucene.472066.n3.nabble.com/distributed-search-is-significantly-slower-than-direct-search-td4098191.html#a4101713
I upload a patch for testing performance.
The idea is during the top document search (first stage) we read the uniquekey from docValues, and doc fields are not loaded (not normally nor lazily). During the second stage we do not access the stored fields in case the user requested uniquekey only

> Speed-up distributed search with high rows param or deep paging by transforming docId's to uniqueKey via memory docValues
> -------------------------------------------------------------------------------------------------------------------------
>
>                 Key: SOLR-5478
>                 URL: https://issues.apache.org/jira/browse/SOLR-5478
>             Project: Solr
>          Issue Type: Improvement
>          Components: Response Writers
>    Affects Versions: 4.5
>            Reporter: Manuel Lenormand
>             Fix For: 4.6
>
>




--
This message was sent by Atlassian JIRA
(v6.1#6144)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org