[
https://issues.apache.org/jira/browse/HAWQ-452?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15214455#comment-15214455
]
ASF GitHub Bot commented on HAWQ-452:
-------------------------------------
Github user gcaragea commented on the pull request:
https://github.com/apache/incubator-hawq/pull/385#issuecomment-202481577
Merged to master: 87d13b673c19d0975cd2408c732c8c4836640e66, closing.
@wengyanqing I think it would be better if you can preserve authorship when
merging PRs. That way the original author can take the credit / blame for a fix
:)
> Remove workfile views from hawq_toolkit
> ---------------------------------------
>
> Key: HAWQ-452
> URL: https://issues.apache.org/jira/browse/HAWQ-452
> Project: Apache HAWQ
> Issue Type: Task
> Components: Query Execution
> Reporter: George Caragea
> Assignee: George Caragea
>
> In the hawq_toolkit schema we have a few views with live information about
> disk space used for workfiles (or spillfiles).
> Generating this views means running a UDF on every segment. This doesn't
> apply to the HAWQ 2 architecture:
> * The notion of "every segment" is not well defined in HAWQ 2, since we have
> virtual segments that are created dynamically based on resource requirements
> and availablitiy
> * There is no mechanism to execute a UDF on all the existing virtual or
> physical segments
> We should remove the views from hawq_toolkit.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
