[
https://issues.apache.org/jira/browse/HADOOP-12924?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15198533#comment-15198533
]
Kai Zheng commented on HADOOP-12924:
------------------------------------
Thanks for the discussions, Zhe and Rui.
bq. Kai mentioned that new RS Java coder and legacy RS coder from HDFS-RAID are
not interoperable. ... they should be treated as 2 different policies rather
than 2 different implementations?
Yes right, but I'm not sure we would have the additional policy for the legacy
coder. I believe the old coder is only useful for migrating old data from old
systems (it's possible because the coder resides in hadoop common side and not
tightly coupled with any HDFS version), but the migrating tool may not likely
go into the current HDFS-EC implementation. So I suggest we still use the
current configuration key for the RS coder, to simplify the current HDFS usage
and also the HH coder implementation. In the feature user guide and doc, we
should have one section about erasure coders and how to configure/use them. It
should explicitly mention the legacy coder isn't compatible with the main RS
coders (new Java coder, and ISA-L based native coder). In fact, anytime users
change the configured codec or coder, they should be aware of the properties of
these coders to understand which one is the right one to be used.
> Add default coder key for creating raw coders
> ---------------------------------------------
>
> Key: HADOOP-12924
> URL: https://issues.apache.org/jira/browse/HADOOP-12924
> Project: Hadoop Common
> Issue Type: Sub-task
> Reporter: Rui Li
> Assignee: Rui Li
> Priority: Minor
> Attachments: HADOOP-12924.1.patch
>
>
> As suggested
> [here|https://issues.apache.org/jira/browse/HADOOP-12826?focusedCommentId=15194402&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15194402].
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
