[
https://issues.apache.org/jira/browse/HDFS-6469?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14020375#comment-14020375
]
Suresh Srinivas commented on HDFS-6469:
---------------------------------------
I had a cursory look at the document. I am very uncomfortable about adding all
this complexity into HDFS. What do others think?
Some very high level comments:
- It would be good to understand the clear advantages and disadvantages
(slowdown due to consensus) of this solution.
- Would be great to know failure handling. For example what happens when a
change cannot be journaled after consensus is reached and memory is updated.
- How are requests from other clients handled when a request is being
processed? If they are blocked, is this not making what is sighted as
bottleneck currently, the namenode single lock, much worser?
I plan on catching up on this proposal completely and respond back early next
week.
A related question. Given your focus on this, are you going to abandon
BackupNode based solutions?
> Coordinated replication of the namespace using ConsensusNode
> ------------------------------------------------------------
>
> Key: HDFS-6469
> URL: https://issues.apache.org/jira/browse/HDFS-6469
> Project: Hadoop HDFS
> Issue Type: New Feature
> Components: namenode
> Affects Versions: 3.0.0
> Reporter: Konstantin Shvachko
> Assignee: Konstantin Shvachko
> Attachments: CNodeDesign.pdf
>
>
> This is a proposal to introduce ConsensusNode - an evolution of the NameNode,
> which enables replication of the namespace on multiple nodes of an HDFS
> cluster by means of a Coordination Engine.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
