[
https://issues.apache.org/jira/browse/HDFS-7716?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14313742#comment-14313742
]
Zhe Zhang commented on HDFS-7716:
---------------------------------
Thanks for the patch Jing! I just finished reading {{BlockInfoStriped}} and it
looks very good overall. Some more detailed comments below:
# {{FSDirectory#addBlock}} causes a minor conflict (method signature) when
applying to current HDFS-7285 branch. I think it's because of the way I
resolved conflicts when rebasing with HDFS-7743 from trunk
# {{BlockInfoStriped#getBlockIndex}} and {{BlockIdManager#convertToGroupID}}
now contain quite similar logic. Might be nice to move the two to the same
place?
# {{removeStorage}} doesn't swap the found storage with the last one, which
makes sense for first {{dataBlockNum + parityBlockNum}} slots. What if we are
removing an over-replicated slot? I guess we should still swap it with the last
node, to avoid growing {{triplets}} too much?
# In {{replaceBlock}}:
{code}
final int size = getCapacity();
newBlockGroup.ensureCapacity(size, false);
{code}
This is to guarantee {{indices}} has the same size as {{triplets}}? When do we
grow {{triplets}} without growing {{indices}}?
> Erasure Coding: extend BlockInfo to handle EC info
> --------------------------------------------------
>
> Key: HDFS-7716
> URL: https://issues.apache.org/jira/browse/HDFS-7716
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Jing Zhao
> Assignee: Jing Zhao
> Attachments: HDFS-7716.000.patch, HDFS-7716.001.patch
>
>
> The current BlockInfo's implementation only supports the replication
> mechanism. To use the same blocksMap handling block group and its data/parity
> blocks, we need to define a new BlockGroupInfo class.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
