[
https://issues.apache.org/jira/browse/HDFS-5134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13759135#comment-13759135
]
Tsz Wo (Nicholas), SZE commented on HDFS-5134:
----------------------------------------------
For the bandwidth for balancer, let's keep it per node since the network
bandwidth of a node depends on the network interface of the node but does not
depend on the number of the storages or the storage types.
> NameNode improvements to per-storage block processing
> -----------------------------------------------------
>
> Key: HDFS-5134
> URL: https://issues.apache.org/jira/browse/HDFS-5134
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: namenode
> Affects Versions: Heterogeneous Storage (HDFS-2832)
> Reporter: Arpit Agarwal
> Assignee: Tsz Wo (Nicholas), SZE
>
> Few tasks out of the NameNode change for multiple storage (HDFS-4987):
> # {{blockContentsStale}} should be per storage.
> # {{DatanodeDescriptor#firstBlockReport}} should also be per storage.
> # {{BlockManager#processReport}} can call {{processFirstBlockReport}} per
> storage now.
> # Synchronization for {{DatanodeDescriptor#numBlocks}} (mentioned by Nicholas
> in 4987).
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
