[jira] [Comment Edited] (HDFS-8391) NN should consider current EC tasks handling count from DN while assigning new tasks

["Yi Liu (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HDFS-8391?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14543190#comment-14543190 ] 

Yi Liu edited comment on HDFS-8391 at 5/14/15 5:00 AM:
-------------------------------------------------------

Uma, yes, I think we should increase xmitsInProgress value for Striped block reconstruction, currently we only do it for continuous block replication, so you can have a patch to increase xmitsInProgress for striped block reconstruction too.
I get your point, ideally striped block reconstruction is more expensive than continuous block replication, I think in current phase, we can use the same configuration for them, and the default value is {{2}} which is small enough.


was (Author: hitliuyi):
Uma, yes, I think we should increase xmitsInProgress value for Striped block reconstruction, currently we only do it for continuous block replication, so you can have a patch to increase xmitsInProgress too.
I get your point, ideally striped block reconstruction is more expensive than continuous block replication, I think in current phase, we can use the same configuration for them, and the default value is {{2}} which is small enough.

> NN should consider current EC tasks handling count from DN while assigning new tasks
> ------------------------------------------------------------------------------------
>
>                 Key: HDFS-8391
>                 URL: https://issues.apache.org/jira/browse/HDFS-8391
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: datanode, namenode
>            Reporter: Uma Maheswara Rao G
>            Assignee: Uma Maheswara Rao G
>
> Currently NN will pick the (maxStreams-XmitsCount) number of ECtasks at a time for assigning to the respective DN.
> {code}
> //get datanode commands
>       final int maxTransfer = blockManager.getMaxReplicationStreams()
>           - xmitsInProgress;
> {code}
> But right now we increment xmitsInProgress count at DN only for regular replication tasks but not for ECWorker tasks.
> So, either we should treat this logic separately for EC or we should consider EC current task handling count form DN.
> This jira for discussing more for this case.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)