[ 
https://issues.apache.org/jira/browse/HDFS-4489?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13642895#comment-13642895
 ] 

Suresh Srinivas commented on HDFS-4489:
---------------------------------------

bq. I think a good performance test is to create a namespace of 150M paths. 
Flood the NN with thousands of concurrent file & directory add/deletes per 
second throughout the namespace. Hopefully there is existing benchmark with 
those properties.
I think we are talking about hashmap entry addition and deletion during adds 
and delete of files, other than increased memory. I am not sure I understand 
the cache pollution part of performance impact, given namenode core objects run 
into GBs in a large setup.

I am currently running some slive tests. But I do not currently have bandwidth 
to setup a namenode with 150M paths (that would require more than 64GB of JVM 
heap). Do you have some bandwidth to do these tests? 
                
> Use InodeID as as an identifier of a file in HDFS protocols and APIs
> --------------------------------------------------------------------
>
>                 Key: HDFS-4489
>                 URL: https://issues.apache.org/jira/browse/HDFS-4489
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: namenode
>            Reporter: Brandon Li
>            Assignee: Brandon Li
>             Fix For: 2.0.5-beta
>
>
> The benefit of using InodeID to uniquely identify a file can be multiple 
> folds. Here are a few of them:
> 1. uniquely identify a file cross rename, related JIRAs include HDFS-4258, 
> HDFS-4437.
> 2. modification checks in tools like distcp. Since a file could have been 
> replaced or renamed to, the file name and size combination is no t reliable, 
> but the combination of file id and size is unique.
> 3. id based protocol support (e.g., NFS)
> 4. to make the pluggable block placement policy use fileid instead of 
> filename (HDFS-385).

--
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

Reply via email to