[
https://issues.apache.org/jira/browse/HDFS-13398?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16441039#comment-16441039
]
Mukul Kumar Singh commented on HDFS-13398:
------------------------------------------
Thanks for working on this [~ajaysachdev]. Please find my comments as following.
1) The current patch is not applying right now. Can you please rebase the patch
over latest trunk ?
2) Also can you please upload the patch with filename as
"HDFS-13398.001.patch". Please follow the guidelines for naming the patch as
https://wiki.apache.org/hadoop/HowToContribute#Naming_your_patch.
3) Also can the config "fs.threads" can be replaced with a command line
argument. I feel that will help in controlling the parallelization for each
command. We can certainly have a default value in the case it is not specified.
4) Also it will be great if some unit tests can be added for the patch.
> Hdfs recursive listing operation is very slow
> ---------------------------------------------
>
> Key: HDFS-13398
> URL: https://issues.apache.org/jira/browse/HDFS-13398
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: hdfs
> Affects Versions: 2.7.1
> Environment: HCFS file system where HDP 2.6.1 is connected to ECS
> (Object Store).
> Reporter: Ajay Sachdev
> Assignee: Ajay Sachdev
> Priority: Major
> Fix For: 2.7.1
>
> Attachments: parallelfsPatch
>
>
> The hdfs dfs -ls -R command is sequential in nature and is very slow for a
> HCFS system. We have seen around 6 mins for 40K directory/files structure.
> The proposal is to use multithreading approach to speed up recursive list, du
> and count operations.
> We have tried a ForkJoinPool implementation to improve performance for
> recursive listing operation.
> [https://github.com/jasoncwik/hadoop-release/tree/parallel-fs-cli]
> commit id :
> 82387c8cd76c2e2761bd7f651122f83d45ae8876
> Another implementation is to use Java Executor Service to improve performance
> to run listing operation in multiple threads in parallel. This has
> significantly reduced the time to 40 secs from 6 mins.
>
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
