[
https://issues.apache.org/jira/browse/HDFS-11394?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15997823#comment-15997823
]
Andrew Wang commented on HDFS-11394:
------------------------------------
Thanks for working on this [~lewuathe]! With this change, there's an additional
~100B per FileStatus, and most of the time it'll be the same EC policy on each
status. For reference, a non-EC filestatus is about 300B, so this is a big
increase.
In HDFS-11565 we used a compact identifier for built-in EC policies. Do you
think we should do something similar to dedupe here, maybe the EC policy name?
> Add method for getting erasure coding policy through WebHDFS
> -------------------------------------------------------------
>
> Key: HDFS-11394
> URL: https://issues.apache.org/jira/browse/HDFS-11394
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: erasure-coding, namenode
> Reporter: Kai Sasaki
> Assignee: Kai Sasaki
> Labels: hdfs-ec-3.0-nice-to-have
> Attachments: HDFS-11394.01.patch, HDFS-11394.02.patch
>
>
> We can expose erasure coding policy by erasure coded directory through
> WebHDFS method as well as storage policy. This information can be used by
> NameNode Web UI and show the detail of erasure coded directories.
> see: HDFS-8196
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
