You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-issues@hadoop.apache.org by "Kai Zheng (JIRA)" <ji...@apache.org> on 2014/11/13 10:08:35 UTC
[jira] [Commented] (HDFS-7345) Local Repairable Codes (LRC)
[ https://issues.apache.org/jira/browse/HDFS-7345?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14209516#comment-14209516 ]
Kai Zheng commented on HDFS-7345:
---------------------------------
To possibly avoid the IP concern, I changed the title and description to reference the more open LRC (Local Repairable Codes). This LRC is different from MS-LRC and has already been implemented by Facebook and Ceph respectively. Note, the concrete algorithms used by Facebook and Ceph are also different, I would document all the differences among the three later to help clarify the situation. With such, I hope this helps the legal staff and community to make the final decision which way should we go.
> Local Repairable Codes (LRC)
> ----------------------------
>
> Key: HDFS-7345
> URL: https://issues.apache.org/jira/browse/HDFS-7345
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Kai Zheng
> Assignee: Kai Zheng
>
> HDFS-7285 proposes to support Erasure Coding inside HDFS, supports multiple Erasure Coding codecs via pluggable codec framework (HDFS-7337) and implements Reed Solomon code by default. This is to support a more advanced coding mechanism, Local Repairable Codes (LRC). As discussed in [Xorbas Hadoop Project Page|http://smahesh.com/HadoopUSC/] and the paper attached there, LRC can reduce the number of erasure coding chunks that need to be read when recovering missing data chunks, while still keeping the storage overhead low.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)