hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Colin Patrick McCabe (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-5052) Add cacheRequest/uncacheRequest support to DFSAdmin and NameNode
Date Thu, 08 Aug 2013 02:57:48 GMT

    [ https://issues.apache.org/jira/browse/HDFS-5052?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13733093#comment-13733093

Colin Patrick McCabe commented on HDFS-5052:

I think the batch API is essential to get good performance when adding or removing a lot of
entries at once.

Re: error returns: I will create a Java object / protobuf for each request, so that we can
add more fields later.  We can add more entries later-- for now it will be pretty simple.

For the error return type-- yes, we could use an integer rather than a string.  That would
be more efficient and make it easier for callers to parse results.  I'll change that.

Cursor-based listing is something we need.  I don't want an API that has problems scaling
up.  We should just do it right the first time around.  I don't think there are any real consistency
issues-- we're not trying to provide ACID transactions here.  I suppose you're right that
RemoteIterator is a smarter way to do it.

Can we keep the edit log stuff in this jira please?  I feel like our subtasks are already
pretty fine-grained, and we have a lot more to go...
> Add cacheRequest/uncacheRequest support to DFSAdmin and NameNode
> ----------------------------------------------------------------
>                 Key: HDFS-5052
>                 URL: https://issues.apache.org/jira/browse/HDFS-5052
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: datanode, namenode
>            Reporter: Colin Patrick McCabe
>            Assignee: Colin Patrick McCabe
>         Attachments: HDFS-5052.003.patch, HDFS-5052.004.patch
> Add cacheRequest/uncacheRequest/listCacheRequest support to DFSAdmin and NameNode.  Maintain
a list of active CachingRequests on the NameNode.

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

View raw message