[
https://issues.apache.org/jira/browse/HDFS-9635?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15096274#comment-15096274
]
Yong Zhang commented on HDFS-9635:
----------------------------------
As guys discussed in HDFS-8538, especially you mentioned
https://issues.apache.org/jira/browse/HDFS-8538?focusedCommentId=14574914&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14574914.
IMO, AvailableSpaceVolumeChoosingPolicy is ok for scenario of disk with
different free space or capacity. we also use
AvailableSpaceVolumeChoosingPolicy by default.
By the way, I don't think collect OS level IO statistics is a good idea.
Considering heterogeneous mechine write performance depends not only disk IO,
but also machine CPU, network bandwith, hadware new/old and so on. So I think
we can collect data writing delay metrics for both BlockPlacementPolicy and
VolumeCoosingPolicy in further work. It is usefull for multi-tenant cluster.
> Add one more volume choosing policy with considering volume IO load
> -------------------------------------------------------------------
>
> Key: HDFS-9635
> URL: https://issues.apache.org/jira/browse/HDFS-9635
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: datanode
> Reporter: Yong Zhang
> Assignee: Yong Zhang
>
> We have RoundRobinVolumeChoosingPolicy and
> AvailableSpaceVolumeChoosingPolicy, but both not consider volume IO load.
> This jira will add a Add one more volume choosing policy base on how many
> xceiver count on volume.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
