[
https://issues.apache.org/jira/browse/HDFS-7717?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14387824#comment-14387824
]
Kai Zheng commented on HDFS-7717:
---------------------------------
bq.Conversion task should be distributed to DNs which uses EC Worker.
Yes, this is confirmed. ECWorker will does the job.
bq.EC Worker design is now updated.
Ah, yes, it's just updated. It's still on going, so please provide your
thoughts.
bq.Conversion tool depends on the conclusive design of EC Worker.
There would be no perfect design for us to depend on. Sure if you have concrete
questions and ideas, let we see how to proceed.
bq.So am I supposed to follow EC Worker design progress and discussion?
Yes, let's discuss first if you have ideas and are going to implement the codes.
> Erasure Coding: distribute replication to erasure coding conversion work to
> DataNode
> ------------------------------------------------------------------------------------
>
> Key: HDFS-7717
> URL: https://issues.apache.org/jira/browse/HDFS-7717
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Jing Zhao
> Assignee: Kai Sasaki
>
> We need some approach to distribute conversion work between replication and
> erasure coding to DataNode. It can be NameNode, or a tool utilizing MR just
> like the current distcp, or another one like the balancer/mover.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
