hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Min Zhou (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HDFS-1052) HDFS scalability with multiple namenodes
Date Mon, 24 May 2010 11:06:30 GMT

    [ https://issues.apache.org/jira/browse/HDFS-1052?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12870569#action_12870569
] 

Min Zhou commented on HDFS-1052:
--------------------------------

I don't think multiple namespaces is a good solution for this issue.  The datasets stored
on our cluster  are shared by many departments of our company. If these datasets are seperated
by a number of namespaces,  there is no data sharing; If  we put them in one namespace managed
by a single NameNode, however, the scalability is limited by NameNode's memory . 
Why don't we employ some distributed metadata management approaches like dynamic subtree patitioning(ceph)
or hash-based partitioning(Lustre) ?

Min


> HDFS scalability with multiple namenodes
> ----------------------------------------
>
>                 Key: HDFS-1052
>                 URL: https://issues.apache.org/jira/browse/HDFS-1052
>             Project: Hadoop HDFS
>          Issue Type: New Feature
>          Components: name-node
>    Affects Versions: 0.22.0
>            Reporter: Suresh Srinivas
>            Assignee: Suresh Srinivas
>         Attachments: Block pool proposal.pdf, Mulitple Namespaces5.pdf
>
>
> HDFS currently uses a single namenode that limits scalability of the cluster. This jira
proposes an architecture to scale the nameservice horizontally using multiple namenodes.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message