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

Posted to issues@kudu.apache.org by "Xu Yao (JIRA)" <ji...@apache.org> on 2018/06/20 03:48:00 UTC

[jira] [Updated] (KUDU-2437) Split a tablet into some chunks by size

     [ https://issues.apache.org/jira/browse/KUDU-2437?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Xu Yao updated KUDU-2437:
-------------------------
    Summary: Split a tablet into some chunks by size  (was: Generate ScanToken from small chunks in tablet)

> Split a tablet into some chunks by size
> ---------------------------------------
>
>                 Key: KUDU-2437
>                 URL: https://issues.apache.org/jira/browse/KUDU-2437
>             Project: Kudu
>          Issue Type: Improvement
>          Components: client, master, tablet
>            Reporter: Xu Yao
>            Assignee: Xu Yao
>            Priority: Major
>
> When reading data in a kudu table using spark, if there is a large amount of data in the tablet, reading the data takes a long time.
> The reason is that KuduRDD uses a tablet to generate the scanToken, so a spark task needs to process all the data in a tablet. So:
>  # TS report the DRS bounds info to Master
>  # Client get the bounds info from Master
>  # Client generate the scanToken by bounds info of tablet（set LowerBoundPrimaryKey and UpperBoundPrimaryKey）



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)