[
https://issues.apache.org/jira/browse/TAJO-1464?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14624102#comment-14624102
]
ASF GitHub Bot commented on TAJO-1464:
--------------------------------------
Github user eminency commented on the pull request:
https://github.com/apache/tajo/pull/579#issuecomment-120780389
Hi, @blrunner
1. Presto-orc is tested and stable. And it is a bit more faster than Hive.
Secondly, Hive processes ORC with row-oriented, but Presto does with
column-oriented. Because ORC is columnar format, Presto's way may be more
suitable. Though Tajo processes it with tuple-oriented, it can be helpful when
Tajo will change to use offheap in the future.
2. Sure.
Anyway, **please hold the review**. About new storage manager code, I'm
modifying.
> Add ORCFileScanner to read ORCFile table
> ----------------------------------------
>
> Key: TAJO-1464
> URL: https://issues.apache.org/jira/browse/TAJO-1464
> Project: Tajo
> Issue Type: Sub-task
> Components: Storage
> Affects Versions: 0.10.0
> Reporter: Dongjoon Hyun
> Assignee: Jongyoung Park
> Fix For: 0.11.0
>
> Attachments: TAJO-1464.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
