hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jing Zhao (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-7937) Erasure Coding: INodeFile quota computation unit tests
Date Mon, 13 Apr 2015 22:29:13 GMT

    [ https://issues.apache.org/jira/browse/HDFS-7937?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14493199#comment-14493199
] 

Jing Zhao commented on HDFS-7937:
---------------------------------

Thanks for working on this, Kai. The patch looks good to me overall. Some further comments:
# The patch needs rebase.
# The changes in ECSchema seem unnecessary
# For some constant numbers in the tests (786433, e.g.), please see if it can be replaced
by some computation using CHUNK size and ECSchema information. Otherwise we can add some javadoc
to explain how it's calculated.
# It looks to me that starting a MiniDFSCluster can be avoided in the test. Can we directly
"new" an INodeFile with striped block feature?

> Erasure Coding: INodeFile quota computation unit tests
> ------------------------------------------------------
>
>                 Key: HDFS-7937
>                 URL: https://issues.apache.org/jira/browse/HDFS-7937
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>            Reporter: Kai Sasaki
>            Assignee: Kai Sasaki
>            Priority: Minor
>         Attachments: HDFS-7937.1.patch, HDFS-7937.2.patch, HDFS-7937.3.patch, HDFS-7937.4.patch,
HDFS-7937.5.patch, HDFS-7937.6.patch
>
>
> Unit test for [HDFS-7826|https://issues.apache.org/jira/browse/HDFS-7826]



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

Mime
View raw message