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] Commented: (HADOOP-1134) Block level CRCs in HDFS
Date Mon, 26 Mar 2007 22:32:32 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-1134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12484249

Raghu Angadi commented on HADOOP-1134:

> I thought we'd keep this at 512B until we'd benchmarked things? 
64KB is only for new blocks. We will keep the 512 byte checksums during the upgrade.

> Aren't we going to simply copy the existing checksums, rather than re-generate new checksums
or verify against existing? That'd be a lot faster. 

You are right. I missed that. will correct it. We will just read the existing CRCs.

>> DN verifies the checksum each time it reads the block data
> I thought that only the client would validate, reporting failures to the DN. Why validate
twice, since we have to validate in the client anyway? 

I am not sure of this. I personally prefer datanode verification (and general ownership of
CRCs). I don't think it adds noticeable latency.. I think CPU overhead would be reasonable.

> 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
>         Assigned To: Raghu Angadi
> 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.

View raw message