You are viewing a plain text version of this content. The canonical link for it is here.

Posted to oak-issues@jackrabbit.apache.org by "Fabrizio Fortino (Jira)" <ji...@apache.org> on 2021/06/28 15:22:00 UTC

[jira] [Resolved] (OAK-9461) Dynamic boost lite

     [ https://issues.apache.org/jira/browse/OAK-9461?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Fabrizio Fortino resolved OAK-9461.
-----------------------------------
    Fix Version/s: 1.42.0
       Resolution: Fixed

> Dynamic boost lite
> ------------------
>
>                 Key: OAK-9461
>                 URL: https://issues.apache.org/jira/browse/OAK-9461
>             Project: Jackrabbit Oak
>          Issue Type: Task
>          Components: indexing
>            Reporter: Fabrizio Fortino
>            Assignee: Fabrizio Fortino
>            Priority: Major
>             Fix For: 1.42.0
>
>
> Dynamic boost uses index time boosting. This requires to have a separate field for each unique term that needs to be boosted. With a high number of terms (thousands), this could result in a sparse index that requires extra disk space. Usually, dynamicBoost and similarityTags are configured on the same field. In this case, similarityTags work on the same set of terms without the need to store boost values. This does not affect the index size.
>  
> With *oak.search.dynamicBoostLite=lucene* no index time boosting is used for lucene types indexes. The terms will affect the query match clause but the scores won't be the same. In summary, in lite mode, the query will have the same recall but lower precision.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)