hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Zhe Zhang (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-8032) Erasure coding: shrink size of indices in BlockInfoStriped
Date Mon, 06 Apr 2015 20:01:12 GMT

    [ https://issues.apache.org/jira/browse/HDFS-8032?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14481828#comment-14481828
] 

Zhe Zhang commented on HDFS-8032:
---------------------------------

Thanks Yi for the comment. I should have put a more clear description. This JIRA is mainly
about cutting the number of indices. We can certainly shrink the size of each index as you
suggested.

> Erasure coding: shrink size of indices in BlockInfoStriped
> ----------------------------------------------------------
>
>                 Key: HDFS-8032
>                 URL: https://issues.apache.org/jira/browse/HDFS-8032
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>            Reporter: Zhe Zhang
>            Assignee: Zhe Zhang
>
> In a _m+k_ erasure coding schema, the first _(m+k)_ slots in {{BlockInfoStriped}} do
not need indexing, since the _ith_ storage directly maps to the _ith_ block in the erasure
coding block group.



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

Mime
View raw message