[jira] [Commented] (HDFS-11715) Ozone: SCM : Add priority for datanode commands

2017-09-06 Thread Weiwei Yang (JIRA)

[ 
https://issues.apache.org/jira/browse/HDFS-11715?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16156469#comment-16156469
 ] 

Weiwei Yang commented on HDFS-11715:


Hi [~anu], I don't think this is required before the merge, this can be a 
post-merge task. Thanks

> Ozone: SCM : Add priority for datanode commands
> ---
>
> Key: HDFS-11715
> URL: https://issues.apache.org/jira/browse/HDFS-11715
> Project: Hadoop HDFS
>  Issue Type: Sub-task
>  Components: ozone
>Reporter: Anu Engineer
>Assignee: Anu Engineer
>  Labels: OzonePostMerge, tocheck
>
> While reviewing HDFS-11493, [~cheersyang] commented that it would be a good 
> idea to support priority for datanode commands send from SCM.
> bq. The queue seems to be time ordered, I think it will be better to support 
> priority as well. Commands may have different priority, for example, 
> replicate a container priority is usually higher than delete a container 
> replica; replicate a container also may have different priorities according 
> to the number of replicas 
> This jira tracks that work item.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

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



[jira] [Commented] (HDFS-11715) Ozone: SCM : Add priority for datanode commands

2017-09-06 Thread Anu Engineer (JIRA)

[ 
https://issues.apache.org/jira/browse/HDFS-11715?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16156163#comment-16156163
 ] 

Anu Engineer commented on HDFS-11715:
-

@weiwei yang, Let us chat about this some time. Should we get this done before 
ozone trunk merge ? 


> Ozone: SCM : Add priority for datanode commands
> ---
>
> Key: HDFS-11715
> URL: https://issues.apache.org/jira/browse/HDFS-11715
> Project: Hadoop HDFS
>  Issue Type: Sub-task
>  Components: ozone
>Reporter: Anu Engineer
>Assignee: Anu Engineer
>  Labels: OzonePostMerge, tocheck
>
> While reviewing HDFS-11493, [~cheersyang] commented that it would be a good 
> idea to support priority for datanode commands send from SCM.
> bq. The queue seems to be time ordered, I think it will be better to support 
> priority as well. Commands may have different priority, for example, 
> replicate a container priority is usually higher than delete a container 
> replica; replicate a container also may have different priorities according 
> to the number of replicas 
> This jira tracks that work item.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

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



[jira] [Commented] (HDFS-11715) Ozone: SCM : Add priority for datanode commands

2017-04-27 Thread Chen Liang (JIRA)

[ 
https://issues.apache.org/jira/browse/HDFS-11715?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15987872#comment-15987872
 ] 

Chen Liang commented on HDFS-11715:
---

Thanks [~anu] for filing this, a very interesting idea!

A side note though, what if we do want to delete first to, say, make some space 
for replicate? and will the delete be starved forever if there is replicate 
consistently coming (may not be a practical scenario though). My concern is 
just that the actual algorithm that prioritize things properly can be very 
tricky.

> Ozone: SCM : Add priority for datanode commands
> ---
>
> Key: HDFS-11715
> URL: https://issues.apache.org/jira/browse/HDFS-11715
> Project: Hadoop HDFS
>  Issue Type: Sub-task
>  Components: ozone
>Reporter: Anu Engineer
>Assignee: Anu Engineer
>
> While reviewing HDFS-11493, [~cheersyang] commented that it would be a good 
> idea to support priority for datanode commands send from SCM.
> bq. The queue seems to be time ordered, I think it will be better to support 
> priority as well. Commands may have different priority, for example, 
> replicate a container priority is usually higher than delete a container 
> replica; replicate a container also may have different priorities according 
> to the number of replicas 
> This jira tracks that work item.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

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