[jira] Commented: (HADOOP-1134) Block level CRCs in HDFS

["Raghu Angadi (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HADOOP-1134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12500508 ] 

Raghu Angadi commented on HADOOP-1134:
--------------------------------------


bq. the FileSystem implementation should also try to arrange to checksum that data as it is delivered to the application from the stream, rather than as it is read into the stream's internal buffer. Do you disagree with this?

I agree. How do we implementation this? My preferred method is to FS implementation to do the the buffering.. implementation can still respect the user supplied bufferSize. This will also fix current problem with current ChecksumFileSystem. Does it make sense? 

Do you want the code sharing mainly in software sense or mainly to fix the buffering issue?





> Block level CRCs in HDFS
> ------------------------
>
>                 Key: HADOOP-1134
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1134
>             Project: Hadoop
>          Issue Type: New Feature
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>         Attachments: bc-no-upgrade-05302007.patch, DfsBlockCrcDesign-05305007.htm
>
>
> Currently CRCs are handled at FileSystem level and are transparent to core HDFS. See recent improvement HADOOP-928 ( that can add checksums to a given filesystem ) regd more about it. Though this served us well there a few disadvantages :
> 1) This doubles namespace in HDFS ( or other filesystem implementations ). In many cases, it nearly doubles the number of blocks. Taking namenode out of CRCs would nearly double namespace performance both in terms of CPU and memory.
> 2) Since CRCs are transparent to HDFS, it can not actively detect corrupted blocks. With block level CRCs, Datanode can periodically verify the checksums and report corruptions to namnode such that name replicas can be created.
> We propose to have CRCs maintained for all HDFS data in much the same way as in GFS. I will update the jira with detailed requirements and design. This will include same guarantees provided by current implementation and will include a upgrade of current data.
>  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.