[ 
https://issues.apache.org/jira/browse/HIVE-2925?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14052991#comment-14052991
 ] 

Lefty Leverenz commented on HIVE-2925:
--------------------------------------

*hive.fetch.task.conversion* is documented in the wiki here, next to the other 
fetch task parameters:

* [Configuration Properties -- hive.fetch.task.conversion | 
https://cwiki.apache.org/confluence/display/Hive/Configuration+Properties#ConfigurationProperties-hive.fetch.task.conversion]

Question 1:  What does RS mean in the parameter description:  "should not have 
any aggregations or distincts (which incurs RS)"?  Is it ReduceSinkOperator as 
in HIVE-5056?

Question 2:  What do "LIMIT only" and "LIMIT only (TABLESAMPLE, virtual 
columns)" mean?  I don't see why the latter is less restrictive than the 
former, even after checking the wiki:  
[https://cwiki.apache.org/confluence/display/Hive/LanguageManual+Select#LanguageManualSelect-LIMITClause].

Question 3:  Does FILTER just mean the WHERE and HAVING clauses?  If it isn't 
syntax, shouldn't it be lowercase in the explanations of minimal and more (and 
STAR be *)?

# minimal: SELECT *, filter on partition columns, LIMIT only
# more: SELECT, filter, LIMIT only (TABLESAMPLE, virtual columns)


> Support non-MR fetching for simple queries with select/limit/filter 
> operations only
> -----------------------------------------------------------------------------------
>
>                 Key: HIVE-2925
>                 URL: https://issues.apache.org/jira/browse/HIVE-2925
>             Project: Hive
>          Issue Type: Improvement
>    Affects Versions: 0.10.0
>            Reporter: Navis
>            Assignee: Navis
>            Priority: Trivial
>             Fix For: 0.10.0
>
>         Attachments: ASF.LICENSE.NOT.GRANTED--HIVE-2925.D2607.1.patch, 
> ASF.LICENSE.NOT.GRANTED--HIVE-2925.D2607.2.patch, 
> ASF.LICENSE.NOT.GRANTED--HIVE-2925.D2607.3.patch, 
> ASF.LICENSE.NOT.GRANTED--HIVE-2925.D2607.4.patch, HIVE-2925.1.patch.txt, 
> HIVE-2925.2.patch.txt, HIVE-2925.3.patch.txt
>
>
> It's trivial but frequently asked by end-users. Currently, select queries 
> with simple conditions or limit should run MR job which takes some time 
> especially for big tables, making the people irritated.
> For that kind of simple queries, using fetch task would make them happy.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Reply via email to