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

Posted to issues@carbondata.apache.org by "Jihong MA (JIRA)" <ji...@apache.org> on 2016/12/16 02:28:58 UTC

[jira] [Updated] (CARBONDATA-437) Optimizing collectData in ScannedResultCollector

     [ https://issues.apache.org/jira/browse/CARBONDATA-437?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jihong MA updated CARBONDATA-437:
---------------------------------
    Description: row assembly speed is slow compared to other columnar file format after file scan.  (was: we compared the record assembly from memory when we already load data from hdfs. For parquet, it takes 43 ms for 100000 records, meanwhile carbon takes 86 ms for 100000 records )

> Optimizing collectData in ScannedResultCollector
> ------------------------------------------------
>
>                 Key: CARBONDATA-437
>                 URL: https://issues.apache.org/jira/browse/CARBONDATA-437
>             Project: CarbonData
>          Issue Type: Sub-task
>            Reporter: suo tong
>            Assignee: Raghunandan S
>
> row assembly speed is slow compared to other columnar file format after file scan.



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