[jira] [Commented] (HIVE-11968) make HS2 core components re-usable across projects

["Thejas M Nair (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HIVE-11968?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14908484#comment-14908484 ] 

Thejas M Nair commented on HIVE-11968:
--------------------------------------

[~cwsteinbach] had an early proposal for this here - https://cwiki.apache.org/confluence/display/Hive/AccessServer+Design+Proposal


> make HS2 core components re-usable across projects
> --------------------------------------------------
>
>                 Key: HIVE-11968
>                 URL: https://issues.apache.org/jira/browse/HIVE-11968
>             Project: Hive
>          Issue Type: Bug
>          Components: HiveServer2
>            Reporter: Thejas M Nair
>
> HS2 provides jdbc and odbc access to hive. There has been a lot of investment into HS2 over time ( Fault tolerance, authentication modes, HTTP transport mode, encryption, delegation tokens .. ).
> The thrift API that HS2 provides is generic and is applicable to other SQL engines as well. Spark is already using a fork of HS2, but as it is a fork, it hard to maintain. 
> HS2 code is not structured to be easily re-used and extended. If we can make improvements there, it can be easily re-used by other projects, and all effort can be combined.



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