Yang Yun commented on HDFS-15028:

I have no number how much the overhead is reduced by the change for the local 
disk. I think it can be ignored .But we met this issue when we mount a slow 
volume from remote as Archive storage. 

Yes, it break the underlying changing of the volume size. the fix should be 
configurable and disabled by default to keep current behavior.

> Keep the capacity of volume and reduce a system call
> ----------------------------------------------------
>                 Key: HDFS-15028
>                 URL: https://issues.apache.org/jira/browse/HDFS-15028
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: datanode
>            Reporter: Yang Yun
>            Assignee: Yang Yun
>            Priority: Minor
>         Attachments: HDFS-15028.patch
> The local volume is not changed. so keep the first value of the capacity and 
> reuse for each heartbeat.

This message was sent by Atlassian Jira

To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org

Reply via email to