[
https://issues.apache.org/jira/browse/HDFS-7912?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14364334#comment-14364334
]
Jing Zhao commented on HDFS-7912:
---------------------------------
Thanks for the review, Zhe!
bq. My only concern is that it adds some getStoredBlock calls
The patch adds {{getStoredBlock}} call in mainly two places:
{{removeStoredBlock}} and {{addBlock}}. The {{addBlock}} function is currently
called to handle a "RECEIVED_BLOCK" msg in incremental block report, and it is
not happening when handling a full block report. Thus I guess here we will not
bring too much extra overhead to the NN.
bq. do we need storedBlock in the following code
Here my thought is that we should use BlockInfo for {{countNodes}}, since the
function tries to retrieve the storages from the blocksMap for the given block.
And with this we need to make the change in NameNodeFsck.
> Erasure Coding: track BlockInfo instead of Block in UnderReplicatedBlocks and
> PendingReplicationBlocks
> ------------------------------------------------------------------------------------------------------
>
> Key: HDFS-7912
> URL: https://issues.apache.org/jira/browse/HDFS-7912
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Jing Zhao
> Assignee: Jing Zhao
> Attachments: HDFS-7912.000.patch
>
>
> Now with striped blocks and the design that uses a single BlockInfoStriped
> object to track all the corresponding blocks, we need to clearly distinguish
> the type Block and BlockInfo in BlockManager. Specifically, data structures
> like {{UnderReplicatedBlocks}} and {{PendingReplicationBlocks}} should track
> BlockInfo instead of Block in order to support striped block recovery.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
