[
https://issues.apache.org/jira/browse/HIVE-15546?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15832644#comment-15832644
]
Sahil Takiar commented on HIVE-15546:
-------------------------------------
[~sershe] yes, when {{Utilities.getInputPaths()}} is called (it is called right
before split generation) the method has to check if each given input path (e.g.
a partition or a table) is empty or not. In order to do this, it has to run a
listStatus against the actual filesystem. If it is not empty, Hive can proceed
normally, if it is empty, then it can use {{NullScanFileSystem}} /
{{ZeroRowsInputFormat}} to add a "dummy" path to the partition / table
description.
In this patch, I just made the listStatus against the filesystem
multi-threaded, since it is unavoidable. Making this listStatus multi-threaded
should provide significant performance gains.
> Optimize Utilities.getInputPaths() so each listStatus of a partition is done
> in parallel
> ----------------------------------------------------------------------------------------
>
> Key: HIVE-15546
> URL: https://issues.apache.org/jira/browse/HIVE-15546
> Project: Hive
> Issue Type: Sub-task
> Components: Hive
> Reporter: Sahil Takiar
> Assignee: Sahil Takiar
> Attachments: HIVE-15546.1.patch, HIVE-15546.2.patch,
> HIVE-15546.3.patch, HIVE-15546.4.patch, HIVE-15546.5.patch
>
>
> When running on blobstores (like S3) where metadata operations (like
> listStatus) are costly, Utilities.getInputPaths() can add significant
> overhead when setting up the input paths for an MR / Spark / Tez job.
> The method performs a listStatus on all input paths in order to check if the
> path is empty. If the path is empty, a dummy file is created for the given
> partition. This is all done sequentially. This can be really slow when there
> are a lot of empty partitions. Even when all partitions have input data, this
> can take a long time.
> We should either:
> (1) Just remove the logic to check if each input path is empty, and handle
> any edge cases accordingly.
> (2) Multi-thread the listStatus calls
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
