You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hive.apache.org by "Zheng Shao (JIRA)" <ji...@apache.org> on 2010/02/12 08:29:27 UTC
[jira] Created: (HIVE-1159) QueryPlan.outputs should be determined
at compile time
QueryPlan.outputs should be determined at compile time
------------------------------------------------------
Key: HIVE-1159
URL: https://issues.apache.org/jira/browse/HIVE-1159
Project: Hadoop Hive
Issue Type: Improvement
Affects Versions: 0.6.0
Reporter: Zheng Shao
Assignee: Zheng Shao
Currently, we change QueryPlan.outputs at execution time of DDLTask, MoveTask. The post execution hook relies on QueryPlan.outputs to log the changed entities.
This makes the serialization hard, because we need to serialize the outputs when each of the Tasks are done. This is even harder given that we allow parallel execution of tasks now.
We should be able to determine all that at compile time.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HIVE-1159) QueryPlan.outputs should be
determined at compile time
Posted by "Zheng Shao (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HIVE-1159?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12837005#action_12837005 ]
Zheng Shao commented on HIVE-1159:
----------------------------------
Actually the changes of outputs only happen to DDL commands.
For normal "insert overwrite", the outputs are determined at compile time.
> QueryPlan.outputs should be determined at compile time
> ------------------------------------------------------
>
> Key: HIVE-1159
> URL: https://issues.apache.org/jira/browse/HIVE-1159
> Project: Hadoop Hive
> Issue Type: Improvement
> Affects Versions: 0.6.0
> Reporter: Zheng Shao
> Assignee: Zheng Shao
>
> Currently, we change QueryPlan.outputs at execution time of DDLTask, MoveTask. The post execution hook relies on QueryPlan.outputs to log the changed entities.
> This makes the serialization hard, because we need to serialize the outputs when each of the Tasks are done. This is even harder given that we allow parallel execution of tasks now.
> We should be able to determine all that at compile time.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.