[
https://issues.apache.org/jira/browse/HADOOP-11828?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15071796#comment-15071796
]
jack liuquan commented on HADOOP-11828:
---------------------------------------
Hi Kai,
* In raw erasure coder level, you can set the {{ALLOW_CHANGE_INPUTS}} coder
option to ensure the input buffers are not changed during encoding/decoding.
Thus in HH coder level, you don't need to clone the input buffers thus avoids
data copy.
When I tested, I found that after running encode() of RS, non-direct buffer
input's position will move to end. but direct buffer input's position will not
move. Is that be OK?
If I don't clone the input buffers in HH coder level, I will deal with
non-direct buffer input and move input postion to the begin after I call
encode() of RS.
> Implement the Hitchhiker erasure coding algorithm
> -------------------------------------------------
>
> Key: HADOOP-11828
> URL: https://issues.apache.org/jira/browse/HADOOP-11828
> Project: Hadoop Common
> Issue Type: Sub-task
> Reporter: Zhe Zhang
> Assignee: jack liuquan
> Attachments: 7715-hitchhikerXOR-v2-testcode.patch,
> 7715-hitchhikerXOR-v2.patch, HADOOP-11828-hitchhikerXOR-V3.patch,
> HADOOP-11828-hitchhikerXOR-V4.patch, HADOOP-11828-hitchhikerXOR-V5.patch,
> HDFS-7715-hhxor-decoder.patch, HDFS-7715-hhxor-encoder.patch
>
>
> [Hitchhiker |
> http://www.eecs.berkeley.edu/~nihar/publications/Hitchhiker_SIGCOMM14.pdf] is
> a new erasure coding algorithm developed as a research project at UC
> Berkeley. It has been shown to reduce network traffic and disk I/O by 25%-45%
> during data reconstruction while retaining the same storage capacity and
> failure tolerance capability as RS codes. This JIRA aims to introduce
> Hitchhiker to the HDFS-EC framework, as one of the pluggable codec algorithms.
> The existing implementation is based on HDFS-RAID.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
