hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Raghu Angadi (JIRA)" <j...@apache.org>
Subject [jira] Updated: (HADOOP-1134) Block level CRCs in HDFS
Date Tue, 03 Jul 2007 23:24:05 GMT

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

Raghu Angadi updated HADOOP-1134:
---------------------------------

    Attachment: BlockLevelCrc-07032007.patch

Attaching  BlockLevelCrc-07032007.patch

This pretty much works. sort with validation passed on 500 nodes. The upgrade works but not
very well tested. This a good patch for review and running a test cluster with fresh data.

To apply this patch, you need to apply patch in HADOOP-1286 first.  This patch does not use
InputChecker yet (HADOOP-1470).

We are still testing this. Haven't done much performance much. Sort on 500 nodes runs at comparable
speed.

This is still work in progress (close to finishing).
Comments are welcome. 

> 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, BlockLevelCrc-07032007.patch, DfsBlockCrcDesign-05305007.htm,
readBuffer.java, readBuffer.java
>
>
> 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.


Mime
View raw message