hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Virajith Jalaparti (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-13283) Percentage based Reserved Space Calculation for DataNode
Date Thu, 15 Mar 2018 00:45:00 GMT

    [ https://issues.apache.org/jira/browse/HDFS-13283?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16399682#comment-16399682

Virajith Jalaparti commented on HDFS-13283:

Thanks for reporting this [~lukmajercak], and posting a patch.

One question -- why use an enum to define this and not directly specify the class to use in
the configuration? I think the latter makes this more general. 
While {{PERCENTAGE}} seems like a simple alternative of the absolute value used before, {{CONSERVATIVE}}
and {{AGGRESSIVE}} seem like something specific someone would want to configure. If someone
wants to define a custom policy for defining the reserved space at a later time, then they
would have to add a new enum. I would rather they just implement a new sub-class of {{ReservedSpaceCalculator}}
and directly configure. This will allow defining and configuring the custom policies easier,
and reduce divergence from trunk.

> Percentage based Reserved Space Calculation for DataNode
> --------------------------------------------------------
>                 Key: HDFS-13283
>                 URL: https://issues.apache.org/jira/browse/HDFS-13283
>             Project: Hadoop HDFS
>          Issue Type: New Feature
>          Components: datanode, hdfs
>            Reporter: Lukas Majercak
>            Assignee: Lukas Majercak
>            Priority: Major
>         Attachments: HDFS-13283.000.patch
> Currently, the only way to configure reserved disk space for non-HDFS data on a DataNode
is a constant value via {{dfs.datanode.du.reserved}}. This can be an issue in non-heterogeneous
clusters where size of DNs can differ. The proposed solution is to allow percentage based
configuration (and their combination):
>  ** based on absolute number of reserved space
>  ** based on percentage of total capacity in the storage
>  ** calculates both of the above and takes the one that will yield more reserved space
>  ** calculates 1. 2. and takes the one that will yield less reserved space

This message was sent by Atlassian JIRA

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

View raw message