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

Posted to issues@kylin.apache.org by "XIE FAN (JIRA)" <ji...@apache.org> on 2016/11/08 01:46:58 UTC

[jira] [Updated] (KYLIN-1851) Improve build dictionary, consider that input is already sorted

     [ https://issues.apache.org/jira/browse/KYLIN-1851?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

XIE FAN updated KYLIN-1851:
---------------------------
    Attachment: 0001-KYLIN-1851-unfinished-add-TrieDictionaryForest-and-N.patch

> Improve build dictionary, consider that input is already sorted
> ---------------------------------------------------------------
>
>                 Key: KYLIN-1851
>                 URL: https://issues.apache.org/jira/browse/KYLIN-1851
>             Project: Kylin
>          Issue Type: Bug
>            Reporter: liyang
>         Attachments: 0001-KYLIN-1851-unfinished-add-TrieDictionaryForest-and-N.patch
>
>
> Currently dictionary build may encounter OOM when cardinality is huge. This can benefit from that the input (which is the output of FactDistinctColumn reducer) is already sorted.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)