[ 
https://issues.apache.org/jira/browse/HADOOP-11828?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15110481#comment-15110481
 ] 

jack liuquan commented on HADOOP-11828:
---------------------------------------

Hi [~zhz],
Thanks for your review!
bq.For an util class with all static methods, we don't need a constructor.
I add a private constructor for checkstyle issue, just reference to code of 
{{DumpUtil}} class in {{rawcoder}}
bq.The getPiggyBacksFromInput method is fairly complex and deserves a unit 
test. A ASCII illustration would also be very helpful, similar to Figure 4 in 
the Hitchhiker paper .
Although {{getPiggyBacksFromInput }} is fairly complex, there only one running 
branch in it. I think current unit test cases are good to cover it. I will add 
a ASCII illustration for it.
bq.3.Maybe I'm missing something, but how do we guarantee the length of inputs 
passed to performCoding is always numDataUnits * subPacketSize?
As Rashmi said, subPacketSize is always 2 in Hitchhiker. I think we can 
guarantee it when we preparing block chunks.

> 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, 
> HADOOP-11828-hitchhikerXOR-V6.patch, HADOOP-11828-hitchhikerXOR-V7.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)

Reply via email to