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-7716) Erasure Coding: extend BlockInfo to handle EC info
Date Thu, 05 Feb 2015 07:44:35 GMT

    [ https://issues.apache.org/jira/browse/HDFS-7716?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14306789#comment-14306789
] 

Jing Zhao commented on HDFS-7716:
---------------------------------

Thanks for the comments, Zhe! I'm currently still working on a patch. For your first two questions,
I still plan to use triplets since currently the block report processing logic depends on
this data structure. The difference between {{BlockReplicationInfo}} and {{BlockGroupInfo}}
mainly lies in that {{BlockGroupInfo}} requires the triplets to be sorted based on the block
sequence in the group. Thus the {{addStorage}}, {{removeStorage}} and {{numNodes}} will be
different in these two classes. In the meanwhile, to handle over-replicated blocks in a block
group (due to temporary datanode failure etc.) we may need extra data structure in {{BlockGroupInfo}}
to explicitly record the block index of each triplet.

> Erasure Coding: extend BlockInfo to handle EC info
> --------------------------------------------------
>
>                 Key: HDFS-7716
>                 URL: https://issues.apache.org/jira/browse/HDFS-7716
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>            Reporter: Jing Zhao
>            Assignee: Jing Zhao
>
> The current BlockInfo's implementation only supports the replication mechanism. To use
the same blocksMap handling block group and its data/parity blocks, we need to define a new
BlockGroupInfo class.



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

Mime
View raw message