[
https://issues.apache.org/jira/browse/HDFS-7613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14540539#comment-14540539
]
Zhe Zhang commented on HDFS-7613:
---------------------------------
Thanks for the feedback [~djp] and sorry about getting to this late.
bq. not sure if BlockPlacementPolicyDefault can meet all cases that
BlockPlacementPolicyEC should be there.
As discussed under HDFS-7891, EC has a higher requirement on rack-aware
distribution, which {{BlockPlacementPolicyDefault}} doesn't quite satisfy.
bq. Also, I see BlockPlacementPolicyEC support rack layer only, do we have plan
to support NodeGroup layer as well? It would be great to make EC can be
suitable for broader scenarios.
[~walter.k.su] might be able to provide more insights. ASAIK we haven't
implemented logic for NodeGroup yet. I think it will be an interesting addition.
Right now we are still focusing on the EC-specific algorithm, and will revisit
the multi-policy proposal a little later.
> Block placement policy for erasure coding groups
> ------------------------------------------------
>
> Key: HDFS-7613
> URL: https://issues.apache.org/jira/browse/HDFS-7613
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Zhe Zhang
> Assignee: Walter Su
> Attachments: HDFS-7613.001.patch
>
>
> Blocks in an erasure coding group should be placed in different failure
> domains -- different DataNodes at the minimum, and different racks ideally.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
