[
https://issues.apache.org/jira/browse/HDFS-6469?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14069598#comment-14069598
]
Konstantin Shvachko commented on HDFS-6469:
-------------------------------------------
??to be practical it needs to be an inline Paxos protocol??
I do remember our conversation on the subject.
* FB Raft implementation is not open source, as far as I know.
* EPaxos is on Github, but a) I don't know if it is ready for production use,
b) I don't know if this community will accept such dependency. BTW it is
written in Go.
* So this leaves me with one available option, which is ZooKeeper.
* Interestingly enough, there is an ongoing effort to extract ZAB library ([see
discussion|https://issues.apache.org/jira/browse/HADOOP-10641?focusedCommentId=14052297&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14052297]
in HADOOP-10641).
* Also the proposal is to make the coordination engine pluggable, so that
people could choose what works better for their production use cases. So all
options above are available.
??comparisons to the topology and windows examples??
I do agree with you these are useful and important features. My intent was not
to say they are not, but rather to give an example that HDFS evolves when
companies and corporations invest in developing features for their "proprietary
extensions", which benefits the community in general via competing or open
source alternatives to such extensions.
> 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)
